WO2004023787A2 - Signal intensity range transformation apparatus and method - Google Patents

Signal intensity range transformation apparatus and method Download PDF

Info

Publication number
WO2004023787A2
WO2004023787A2 PCT/US2003/028028 US0328028W WO2004023787A2 WO 2004023787 A2 WO2004023787 A2 WO 2004023787A2 US 0328028 W US0328028 W US 0328028W WO 2004023787 A2 WO2004023787 A2 WO 2004023787A2
Authority
WO
WIPO (PCT)
Prior art keywords
grey
levels
output
generating
image input
Prior art date
Application number
PCT/US2003/028028
Other languages
French (fr)
Other versions
WO2004023787A3 (en
Inventor
Stephen G. Johnson
Norbert W. Elsdoerfer
Charles M. Anastasia
Original Assignee
Rytec Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Rytec Corporation filed Critical Rytec Corporation
Priority to AU2003270386A priority Critical patent/AU2003270386A1/en
Publication of WO2004023787A2 publication Critical patent/WO2004023787A2/en
Publication of WO2004023787A3 publication Critical patent/WO2004023787A3/en

Links

Classifications

    • G06T5/70
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/40Image enhancement or restoration by the use of histogram techniques
    • G06T5/94
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/14Picture signal circuitry for video frequency region
    • H04N5/20Circuitry for controlling amplitude response
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10016Video; Image sequence
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20024Filtering details
    • G06T2207/20032Median filtering
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00Details of colour television systems
    • H04N9/64Circuits for processing colour signals

Definitions

  • the present invention generally relates to imaging systems, and in particular, to a system for manipulating image data.
  • Extreme lighting variation for example, due to sunlight beams, can cause typical video cameras and imaging sensors to saturate (i.e., become unable to represent the real- world luminance range) resulting in wide-scale bright and dark regions having extremely low-contrast wherein objects are difficult or impossible to discern.
  • sunlight beams or strong lighting in the background can cause a person in a dark area to become unidentifiable due to low contrast. This weakness is due to the limited luminance range of the imaging system.
  • the electronic iris and automatic gain control provided by some imaging systems are designed to try to optimally map the wide luminance values within a real- world light situation into a limited range digital representation, often resulting in a poor compromise. To adequately represent the bright areas, less illuminated areas become dramatically compressed in contrast and thus become very dark.
  • video or imagery from lower-cost imaging sensors can have significant noise due to a number of basic system limitations, as well as significant blur due to lower-cost, small, or simple optical configurations. Reducing noise and blur within these systems can improve the ability for a human viewer to effectively interpret image content.
  • a system and method has been developed that considers the mechanisms and defects of imagery from video or other sources to prescribe a sequence of tailored image processing operations designed to improve human interpretation of resulting images.
  • methods and devices are provided to redistribute discrete signal intensity values within groups of signal intensity values.
  • the signal intensity values are supplied directly or indirectly from a sensor sensing an environment.
  • the inventions can be used for enhanced interpretation of any array of signal intensities or variable values in any group of such values that have spatial or geometric relationships to one another (e.g. coordinates).
  • the detailed disclosures below are directed to redistribution of grey-scale or luminance values in a rectilinear array (pixels) from a camera. It is also contemplated however, that the inventions may be applied to other values in a video image, such as redistribution of chrominance values.
  • the inventions may be employed with other sensing modalities such as, magnetic resonance imagery, radar, sonar, infrared, ultraviolet, microwave, X-ray, radio wave, and the like.
  • the spatial scale of an entire group of signal intensity values are considered, for example, the luminance in an entire pixel image, so that the overall brightness from corner to corner is taken into account.
  • all the frequency content of an image must be represented by the one global mapping, including the very lowest spatial frequency.
  • this single global mapping has to stretch to map all of the low frequency variation in the image, and then fails to enhance higher frequency structure of interest. The result can be a bright "bloom" on one side of the image, with too dark an area on the other side of the image. As such, there may not be optimal recognition of spatial structure because of the need to represent the large scale variation across the entire image.
  • the inventions propose generating subset partitions of an image (group of signal intensities) representative of the spatial scales of interest.
  • a transform mapping e.g. of luminance
  • a spatial scale e.g., is 1/4 of a global image for a quadrant subset
  • the present invention employs accumulating pixel samples across larger "right-sized" spatial scales, while using efficient interpolation to produce the correct transform (e.g. for luminance) representation at each pixel.
  • apparatus and methods include: decomposing a group of signal values into subgroup partitions from which to sample values and construct associated transforms and to combine those transform values at every coordinate (e.g. pixel) according to a rule which weights the contribution of each mapping in accordance with "image" geometry and value coordinate or position.
  • apparatus and methods are provided for blending transformed values from the global set of values (e.g. entire pixel image) with transformed values from the spatial segments to adjust contributions from several spatial scales as desired.
  • preferred operations can include digital sampling and luminance/color separation, noise reduction, deblurring, pixel noise reduction, histogram smoothing, contrast stretching, and luminance and chrominance re-combination.
  • One or more of the operators can have configurable attributes, such as degree of noise reduction, brightness, degree of deblurring, and determined range of useful grey-levels.
  • FIGURE 1 is a simplified block diagram of a device in accordance with the present invention, including a logical device;
  • FIGURE 2 is a simplified functional block diagram of a process performed by the logical device of FIGURE 1 , the process having a color transform, an image pre-contrast conditioner, a contrast enhancement, an image post-contrast conditioner, and a color inverse transform;
  • FIGURE 3 is a simplified functional block diagram of the image pre-contrast conditioner of FIGURE 2, the pre-contrast conditioner comprising a system noise reduction filter, a deblurring module, and a pixel noise reduction module;
  • FIGURES 4(a)-(f) depict various exemplary kernels shapes that can be used with the system noise reduction filter, deblurring module, and pixel noise reduction module of FIGURE 3;
  • FIGURE 5 is a simplified functional block diagram of the contrast enhancement block of FIGURE 2, the contrast enhancement comprising an equalized lookup table construction block and an enhanced luminance image generation block;
  • FIGURE 6 is a simplified functional block diagram of the equalized lookup table construction block of FIGURE 5;
  • FIGURE 7 is a simplified functional block diagram of another embodiment of the contrast enhancement block of FIGURE 2.
  • FIGURES 8 and 9 is a simplified function block diagram of yet another embodiment of the contrast enhancement block of FIGURE 2. Detailed Description
  • each of the FIGURES depicts a simplified block diagram wherein each block provides hardware (i.e., circuitry), firmware, software, or any combination thereof that performs one or more operations.
  • hardware i.e., circuitry
  • firmware software
  • Each block can be self-contained or integral with other hardware, firmware, or software associated with one or more other blocks.
  • a device 10 for enhancing, through transformation, the luminance range of an image input signal.
  • the device 10 includes an input connector 12, logic circuitry 14, and an output connector 16.
  • the connectors 12 and 16 are mounted in a conventional manner to an enclosed housing 13, constructed of a metal, metal alloy, rigid plastic, or combinations of the above, that contains the logic circuitry 14.
  • one or more of the modules described herein are performed by the logic circuitry 14 comprising of one or more integrated circuits, commonly referred to as "ICs,” placed on one or more printed circuit boards mounted within the housing 13.
  • ICs integrated circuits
  • the device 10 is a stand-alone or embedded system.
  • the term "stand-alone” refers to a device that is self-contained, one that does not require any other devices to perform its primary functions.
  • a fax machine is a stand-alone device because it does not require a computer, printer, modem, or other device.
  • the device 10 does not need to provide ports for connecting a disk drive, display screen, or a keyboard.
  • the device 10 could provide one or more ports (e.g., RS-232) for supporting field interactivity.
  • an embedded system is a system that is not a desktop computer or a workstation computer or a mainframe computer designed to admit facile human interactivity.
  • Another delineator between embedded and “desktop” systems is that desktop systems (and workstation, etc.) present the status of the computer state to the human operator via a display screen and the internal state of the computer is represented by icons on the screen, and thus the person can interact with the computer internal state via control of the icons.
  • desktop systems and workstation, etc.
  • Such a computer uses a software layer called an "operating system” through which the human operator can interact with the internal state of the computer.
  • an embedded system while it is performing its work function, the human operator cannot interact with the work process except to stop it.
  • the input connector 12 provides for operably connecting the device 10 to an image input signal 17 generated by a video camera (not shown), or the like, having a video output.
  • the input connector 12 consists of an F connector, BNC connector, RCA jacks, or the like.
  • the input connector 12 is operably connected to the logic circuitry 14 by way of a conductive path attached to the input connector and the printed circuit board contained within the housing 13.
  • the logic circuitry could also be coupled through other than a conductive path such as through optical coupling.
  • the image input signal 17 is a conventional analog video signal containing a plurality of still images or fixed image frames taken in a sequential manner.
  • Each frame provided by the image input signal is also referred to herein as an image input frame.
  • Each image or frame includes data regarding an array of pixels contained therein.
  • the output connector 16 of the device 10 provides for connecting the device to an output device such as a monitor (not shown). Like the input connector 12, the output connector 16 consists of any means for outputting the signal to other devices such as, an F connector, BNC connector, RCA jacks, or the like.
  • the output connector 16 is operably connected to the logic circuitry 14 by way of a conductive or coupled path attached to the output connector and the printed circuit board contained within the housing 13.
  • the output signal provided by connector 16, and thus the device 10 provides an output signal which includes data resulting from transforming or other operations carried out with respect to image input signal 17 received ("transformed output signal").
  • the output signal can include a plurality of image output frames and be formatted as a conventional analog video signal, a digital signal, or the like.
  • the output signal can be in a format as defined by NTSC, VGA, HDTV, or other desired output formats.
  • the logic circuitry 14 within the device 10 includes, inter alia, circuitry configured to transform the variable range of grey-scale values in the image input signal 17 received by the input connector 12.
  • the logic circuitry 14 includes a logical device 18 with corresponding support circuitry 20, a video decoder 22, and a video encoder 23.
  • the support circuitry 20 preferably includes a microcontroller 24, a read only memory 26, and a random access memory 28 comprising a synchronous dynamic random access memory.
  • an optional switch 29 is provided for configuring the logic circuitry 14 within the device 10.
  • the switch 29 is operably connected to the microcontroller 24 and logical device 18.
  • the switch 29 allows a user to enable or disable, features or processes provided by the logic circuitry 14 within the device 10.
  • the switch 29 consists of a conventional DIP switch.
  • the configuration of the circuitry 14 within the device 10 is hardwired, or can be set via software commands, instead of using a switch.
  • video decoder 22 is operably connected to the input connector 12 and the logical device 18. Accordingly, the video decoder 22 receives the image input signal 17 that can consist of live video from a television broadcast, a video tape, a camera, or any other desired signals containing or representing image content.
  • the video decoder 22 preferably is a conventional device for tracking the video image input signal 17, digitizing the input signal (if required), separating out the brightness and color information from the input signal, and forwarding the digital video signal 30 to the logical device 18 on a frame by frame basis.
  • the input image signal 17 received by the video decoder 22 is an analog signal formatted in a predefined manner such as PAL, NTSC, or another conventional format.
  • the video decoder 22 converts the analog signal 17 into a digital video signal 30 using a conventional analog-to-digital conversion algorithm .
  • the digital video signal 30 provided by the video decoder 22 includes luminance information and color information in any conventional manner such as, specified by YUV format, YCbCr format, super video, S-video, or the like.
  • the digital video signal 30 can have the luminance information embedded therein such as that provided by digital RGB, for example.
  • the video decoder 22 is capable of converting a plurality of different analog video formats into digital video signals suitable for processing by the logical device 18 as described in detail further herein.
  • the microcontroller 24 configures the video decoder 22 for converting the image input signal
  • the microcontroller 24 determines the format of the image input signal 17, and configures the video decoder 22 accordingly. The determination can be accomplished by the microcontroller 24 checking the user or device manufacturer configured settings of the DIP switch corresponding with the format of the image input signal 17 expected to be received.
  • the video decoder 22 can include circuitry for automatically detecting the format of the image input signal 17, instead of using preset DIP switch settings.
  • the gain of the video decoder 22 is set to reduce overall contrast on the luminance for reducing the probability of image saturation.
  • the video decoder 22 provides a fixed bit resolution output range (e.g., 8 bit, 16 bit, 32 bit, etc.) and a digitizer maps the image input signal in a conventional manner for effective use of the resolution output range.
  • the full range of the digitizer output is utilized.
  • the logical device receives digital video signals 30 and provides digital output signals 31 (i.e., transformed pixel array or frame data) in response to the digital video signals 30.
  • the device 10 receives analog image input signals 17 that are converted by the video decoder 22 into digital video signals 30 for manipulation by the logical device 18. Moreover, as explained in detail further herein, the digital output 31 of the logical device 18 can be converted (if desired) into an analog output by the video encoder 23 connected between the logical device 18 and the output connector 16.
  • the logical device 18 consists of a conventional field programmable gate array (FPGA).
  • the logical device 18 can consists of a Digital Signal Processor (DSP), a microcontroller, an Application Specific Integrated Circuit (ASIC), or other suitable device.
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • the device 10 instead of receiving analog input signals 17, can receive digital image input signals that are provided, without any conversion, to the logical device 18.
  • the video decoder 22 can be omitted.
  • the logical device 18 is operably connected between connectors 12 and
  • an image input signal enters the input connector 12, is modified by the logical device 18, and then exits via the output connector 16 as a transformed output signal.
  • the frame output rate of the device 10 is substantially equal to the frame input rate to the device.
  • the device 10 provides an output of thirty frames per second in response to a frame input rate of thirty frames per second.
  • the logical device 18 provides a sequence of image processing functional modules or blocks 32 within a process as illustrated in FIGURE 2.
  • the blocks 32 represent a color transform 34, an image pre- contrast conditioner 36, a contrast enhancement 38, an image post-contrast conditioner 40, and a color inverse transform 42.
  • Each of the modules 32 preferably performs a specific functional step or plurality of functional steps as described in detail further herein.
  • static data for configuring the logical device 18, if needed or desired is stored within the read only memory 26 operably connected thereto.
  • the read only memory 26 provides for storing data that is not alterable by computer instructions.
  • SDRAM synchronous dynamic random access memory
  • a high speed random access memory is provided by the support circuitry 20 to reduce performance bottlenecks.
  • the memory 20 is used as a field buffer.
  • the color transform 34 provides for separating the luminance from the digital video signal 30, as desired.
  • the digital video signal 30 provided by the video decoder 22 consists of a digital RGB signal.
  • luminance information is not separated from color information.
  • the color transform 34 provides for separating the luminance information for each pixel, within each frame, from the digital video signal 30.
  • the color transform 34 uses a conventional algorithm for converting a digital RGB video input signal into a Yuv format signal, YCbCr format signal, or other desired format signal. Accordingly, the color transform 34 provides an output comprising three channels: luminance 43, and U-V or other values ascribed per pixel location.
  • the digital video signal 30 provided by the video decoder 22 consists of a super video or S-video.
  • the digital video signal 30 consists of two different signals: chrominance and luminance. Accordingly, the color transform 34 can be omitted from the process of FIGURE 2 because the luminance 43 is separately provided by the S-video input.
  • the luminance information 43 contained within the digital video signal 30 is received by the image pre-contrast conditioner 36.
  • the image pre-contrast conditioner 36 can include a video time integration module or block 44, a system noise reduction module or block 46, a deblurring module or block 48, and a pixel noise reduction module or block 50.
  • any one or all of the blocks within FIGURE 3 can be omitted including, but not limited to, the video time integration module 44.
  • the video time integration module 44 consists of a low-pass filter for noise reduction of the luminance signal 43.
  • each frame includes an array of pixels wherein each pixel has a specific i,j coordinate or address.
  • Noise reduction by the video time integration module 44 is preferably achieved by combining the current frame (n) with the immediately preceding frame (n-1) stored by the logical device 18 within the memory 28.
  • the corresponding pixel p(ij) from the preceding captured frame (n- 1 ) is subtracted to result in a difference d(i,j).
  • the absolute value of the difference d(ij) is determined, and compared to a predetermined threshold value. If the absolute value of the difference d(i,j) is greater than the predetermined threshold, then the time integration output signal 45 of the video time integration module 44 is c(i j). Otherwise, the time integration output signal 45 of the video time integration module 44 is the average of c(ij) and p(i,j).
  • the video time integration module 44 can operate such that object motion, which ordinarily creates large pixel differences, are represented by current pixels while background differences due to noise are suppressed by the averaging. Accordingly, the video time integration module 44 provides for avoiding image blurring as a result of low-pass spatial domain filtering on the current image.
  • the median filter or system noise reduction module 46 receives the time integration output signal 45 (i.e., transformed pixel frame data) and, in response thereto, provides for reducing pixel noise caused by the system senor (e.g., camera), electromagnetic, and/or thermal noise.
  • the median filter 46 can include a user selectable level of noise reduction by selecting from a plurality of filter kernels via DIP switch 29, or the like, operably coupled to the logical device 18.
  • the filter kernel applied by the noise reduction module 40 to the current frame, as preferably modified by time integration module 44, is designed to achieve noise reduction while minimizing the adverse effect of the filter on perceived image quality.
  • a 3X1 kernel (FIGURE 4(a)) provides for a reduction in row noise due to the kernel shape, while having minimal adverse effect on the image.
  • a 5-point or plus kernel (FIGURE 4(b)) offers symmetric noise reduction with low adverse impact to image quality.
  • a hybrid median filter (FIGURE 4(c)) offers symmetric noise reduction with lower adverse impact than the full 3X3 kernel.
  • a user or the manufacturer can select from the kernel shapes shown in FIGURES 4(a)-(c).
  • the deblurring module or block 48 provides for counteracting image blurring attributable to the point spread function of the imaging system (not shown).
  • the deblurring module or block 48 uses a Laplacian filter to sharpen the output 47 (i.e., transformed pixel frame data) received from the filter 46.
  • the deblurring module or block 48 can include a user-selectable level of image sharpening from a plurality of Laplacian filter center pixel weights.
  • the user or manufacturer can select the center pixel weight via the DIP switch 29 operably coupled to the logical device 18.
  • the Laplacian filter uses a 3X3 kernel (FIGURE 4(d)) with selectable center pixel weights of 9, 10, 12, or 16, requiring normalization divisors of 1, 2, 4, or 8, respectively.
  • the remaining kernel pixels are preferably assigned a weight of -1.
  • the result of applying the kernel is normalized by dividing the convolution result for the current pixel element by the normalization divisor.
  • the deblurring module or block 48 can use other filters in place of a Laplacian type filter.
  • a general filter with kernel as shown in FIGURE 4(e) can be used for deblurring.
  • the output 49 (i.e., transformed pixel frame data) of the deblurring module 48 emphasizes isolated pixel noise present in the input image signal.
  • the pixel noise reduction module 50 provides for reducing isolated pixel noise emphasized within the output 49 of the deblurring module 48.
  • the pixel noise reduction module 50 employs a small-kernel median filter (i.e., small pixel neighborhood such as a 3X3) to reduce isolated pixel noise.
  • the pixel noise reduction filter or module 50 can provide a user or manufacturer selectable level of noise reduction by allowing for the selection from a plurality of filter kernels.
  • the contrast enhancement module or block 38 enhances the contrast of the conditioned luminance signal 53 (i.e., transformed pixel frame data) provided by the image pre-contrast conditioner 36 in response to the original luminance input signal 17.
  • the contrast enhancement module 38 directly receives the image input signal 17, and in response thereto, generates an enhanced image signal 55.
  • the contrast enhancement module or block 38 preferably includes a sample area selection module or block 52, an equalized lookup table construction module or block 54, and an enhanced luminance generation module or block 56.
  • lookup table and “equalized lookup table” both refer to a lookup table derived from grey-level values or a lookup table representative of the inversion of the cumulative distribution function derived from an accumulated histogram.
  • the sample area selection module or block 52 receives the conditioned luminance image signal 53 provided by the image pre-contrast conditioner 36 (FIGURES 2 and 3) and a selected portion of the input image, via input 57, to be enhanced.
  • the sample area selection module provides a selected conditioned luminance image signal 59 comprising conditioned luminance pixel data from the pixels within the selected region.
  • the selected portion of the image can be selected in a conventional manner such as by using a point-and-click interface (e.g., a mouse) to select a rectangular sub-image of an image displayed on a monitor (not shown).
  • a point-and-click interface e.g., a mouse
  • the selected portion of the image can be selected by automated means such as by a computer, based upon movement detection, with a portion of the image or other criteria.
  • the equalized look-up table construction module or block 54 receives the selected conditioned image signal 59 from the sample area selection module or block 52 and, in response thereto, provides for creating a lookup table 61 that is used in generating the enhanced image signal 55 (i.e., transformed pixel frame data).
  • the selected conditioned luminance signal 59 received from the sample area selection module or block 52 can be a sub-image comprising a portion of the overall conditioned luminance image signal 53 received from the image pre-contrast conditioner 36 (FIGURES 2 and 3).
  • the selected conditioned luminance signal 59 can consist of the complete conditioned luminance signal 53, without any apportioning by the sample area selection module 52.
  • the equalized lookup table construction module, or block 54 of FIGURE 5 includes an input smoothing module or block 58, a histogram accumulation module or block 60, a histogram smoothing module or block 62, a cumulative distribution function integration module or block 64, a saturation bias removal module or block 66, a linear cumulative distribution function scaling module or block 68, and a lookup table construction module or block 70.
  • the input image smoothing module, or block 58 receives the selected conditioned luminance image signal 59 and, in response thereto, generates a smoothed conditioned luminance image signal on a frame-by-frame basis.
  • the input image smoothing module or block 58 applies a Gaussian filter with a traditional 3X3 kernel to the selected signal 59 for smoothing the image within each frame.
  • a histogram accumulation module, or block 60 provides for generating a histogram of the selected conditioned image signal 59, as modified by the input image smoothing module.
  • the histogram accumulation module, or block 60 accumulates a conventional histogram from the received input. Accordingly, the histogram can be used to provide a graph of luminance levels to the number of pixels at each luminance level. Stated another way, the histogram reflects the relationship between luminance levels and the number of pixels at each luminance level.
  • the histogram accumulation module or block 60 includes a plurality of data storage locations or "bins" for tracking the number of occurrences of each grey-level occurring in the received input signal, for example, such as that received by block 58.
  • the histogram accumulation module or block 60 includes a bin for every discrete grey-level in the input signal.
  • the histogram accumulation module or block 60 determines the number of pixels in the input luminance image with the corresponding grey-level after being filtered, if desired. Accordingly, the histogram result for any given grey-level k is the number of pixels in the luminance image input having that grey-level.
  • the histogram smoothing module or block 62 provides for reducing noise in the histogram created from the input image.
  • the histogram smoothing module, or block 62 receives the histogram from the histogram accumulation, module or block 60, and in response thereto, generates a smoothed histogram.
  • the histogram smoothing module or block 62 applies a 5-point symmetric kernel (FIGURE 4(f)) to filter the histogram calculated for each frame.
  • the cumulative distribution function integration module or block 64 provides for integrating the histogram to create a cumulative distribution function.
  • the cumulative distribution function integration module or block 64 receives the histogram from the histogram smoothing module 62, or optionally from the histogram accumulation module 60.
  • the cumulative distribution function integration module 64 integrates the received histogram to generate a cumulative distribution function. It has been observed that integrating a smoothed histogram can result in a cumulative distribution function having an increased accuracy over a cumulative distribution function resulting from an unsmoothed histogram.
  • the cumulative distribution function integration module or block 64 includes a plurality of data storage locations or "bins" which hold the cumulative distribution function result for each input grey- level.
  • the cumulative distribution function integration module, or block 64 includes a bin for every discrete grey-level in the image resolution input range. For each grey-level in the image resolution input range, the cumulative distribution function integration module or block 64 determines the cumulative distribution function result and stores that result in the bin corresponding to that grey-level. For each grey-level k, the cumulative distribution function result is the sum of the histogram value for grey-level k and the cumulative distribution function result for the previous grey-level k- 1. Accordingly, the following equation describes the cumulative distribution function result for a given grey- level k:
  • CDF(k) H(k) + CDF(k-l)
  • CDF(k) is the cumulative distribution function result for grey-level k
  • H(k) is the histogram value for grey-level k
  • CDF(k-l) is the cumulative distribution function result for the grey-level one increment lower than grey-level k.
  • the saturation bias identification module provides for identifying the grey-levels that bound the useful information in the luminance signal 43.
  • the saturation bias identification module 66 receives the cumulative distribution function from the cumulative distribution function integration module 64 and determines the grey-levels at which the first unsaturated grey-level, k f , and the last unsaturated grey-level, k,, occurs.
  • the first unsaturated grey-level, k f is determined by identifying the first grey-level ko for which the cumulative distribution function returns a non-zero value.
  • the grey-level ko is treated as saturated, and the saturation bias identification module 66 identifies k f as the sum of ko plus one additional grey-level.
  • the last unsaturated grey-level, k b is determined by identifying the first grey-level k hoid for which the cumulative distribution function returns the number of pixels in the image.
  • the grey-level k n is treated as saturated, and the saturation bias identification module or block 66 identifies k, as the difference between k navig minus one additional grey-level.
  • the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, kg plus one additional grey- level through k n minus one additional grey-level.
  • the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, ko through k n minus one additional grey-level.
  • the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including ko plus one additional grey-level through k n .
  • the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, ko plus X additional grey-level(s) through k n minus Y additional grey-level(s), wherein X and Y are whole numbers greater than zero.
  • the linear cumulative distribution function scaling module or block 68 provides for scaling the portion of the cumulative distribution function corresponding to the unsaturated, useful grey-levels in the luminance input image across the entire range of cumulative distribution function grey-level inputs.
  • the linear cumulative distribution function scaling module 68 receives the cumulative distribution function, provided by the cumulative distribution function integration module 64, and the first unsaturated grey-level k f and the last unsaturated grey-level k, from the saturation bias identification module 66.
  • the linear cumulative distribution function scaling module 68 includes a plurality of data storage locations or "bins" equal to the number of bins in the cumulative distribution function for holding the linearly mapped cumulative distribution function result for each input grey-level.
  • the cumulative distribution function scaling module 68 scales the portion of the cumulative distribution function between k f and k, inclusively across the entire range of the linearly scaled cumulative distribution function. Accordingly, for each grey-level k in the input range, the linearly scaled cumulative distribution function output value LCDF(k) is determined by the following equation:
  • LCDF(k) (CDF(k) - CDF(kf))/(CDF(Kl) - CDF(Kf))
  • CDF(k) is the cumulative distribution function result for grey-level k
  • CDF(kf) is the cumulative distribution function result for kf
  • CDF(kl) is the cumulative distribution function result for kl.
  • Each linearly mapped cumulative distribution function result is stored in the bin corresponding to the current grey-level k. If LCDF(k) is negative for any input grey-level k, the linearly scaled cumulative distribution function result in the bin corresponding to the input grey-level k is set to zero.
  • the determination of the cumulative distribution function output value can be calculated using known methodologies for improving computation ease.
  • the numerator can be scaled by a multiplier, before the division operation, to obtain proper integer representation.
  • the result of the division can then be inversely scaled by the same multiplier to produce the LCDF(k) value.
  • the scale factor is an adequately large constant, such as the number of pixels in the input image.
  • the lookup table construction module or block 70 generates a lookup table 61 that is used by the enhanced luminance image generation module (FIGURE 5) to produce the enhanced luminance image signal 55.
  • the lookup table construction module 70 receives the linearly scaled cumulative distribution function from the linear cumulative distribution function scaling module 70. In response to the scaled cumulative distribution function, the lookup table construction module 70 provides the lookup table 61.
  • the lookup table 61 is constructed by multiplying each linearly scaled cumulative distribution function result by the number of discrete grey-levels in the output image resolution range which can be provided by the DIP switch 29 (FIGURE 1), software, or other desired means. If the multiplication result is greater than the maximum value in the output image resolution range for any input grey-level, the linearly scaled cumulative distribution function result for that input grey-level is set to the maximum value in the output image resolution range. Accordingly, for each grey- level input in the linearly scaled cumulative distribution function, the result is a value in the output image resolution range.
  • the enhanced luminance image generation module, or block 56 is responsive to the lookup table 61 and the conditioned luminance signal 53.
  • the luminance of each pixel within each input frame can be reassigned a different value based upon the lookup table corresponding to the same frame. Stated another way, the luminance value for each pixel in the conditioned luminance input image is used as an index in the look-up table, and the value at that index location is used as the luminance value for the corresponding pixel in the enhanced luminance output image.
  • the reassigned frame data is then forwarded, preferably to image post contrast conditioner 40, as the enhanced image signal 55.
  • the image post contrast conditioner 40 provides for filtering the enhanced image signal 55 to generate an enhanced conditioned image signal
  • the conditioner 40 can apply to the image signal 55 one or more conventional imaging filters, including an averaging filter, a Gaussian filter, a median filter, a Laplace filter, a sharpen filter, a Sobel filter, and a Prewitt filter.
  • the color inverse transform 42 generates the contrast enhanced digital video output signal 31 , having an expanded range, in response to the enhanced image signal 72 and the color information for the same frame.
  • the enhanced image signal and color information are combined in a conventional manner wherein luminance is combined with color information.
  • the color information can be stored within memory 28, for example. Once the luminance has been enhanced, it can be re-combined with the color information stored in memory.
  • the video encoder 23 preferably provides an analog output signal via output connector 16. The video encoder 23 preferably utilizes conventional circuitry for converting digital video signals into corresponding analog video signals.
  • FIGURE 7 an alternate embodiment of the contrast enhancement module is depicted. Within FIGURE 7, that last two digits of the reference numbers used therein correspond to like two digit reference numbers used for like elements in
  • FIGURES 5 and 6 Accordingly, no further description of these elements is provided.
  • the median bin identification module or block 174 determines the median of the histogram provided by the histogram accumulation module 160.
  • the brightness midpoint value is calculated by the weighted midpoint calculation module or block 176.
  • the brightness midpoint value preferably is calculated as the weighted sum of the actual grey-level midpoint of the input frame and the median of the input distribution.
  • the weights assigned to the grey-level midpoint and the median of the input distribution are configured as compile-time parameters.
  • the image segmentation module, or block 178 receives the luminance image data for the selected sample area, via module 152, and provides for segmenting the luminance image data into light and dark regimes and performing enhancement operations separately (i.e., luminance redistribution) on each of the regimes. Stated another way, the image segmentation module 178 assigns each pixel in the selected sample area to either the light or dark regime based on each pixel's relationship to the brightness midpoint value received from the midpoint calculation module 176. Pixels with a brightness value greater than the midpoint are assigned to the light regime, and pixels with a brightness value less than the midpoint value are assigned to the dark regime.
  • Pixel values assigned by the image segmentation module 178 are received by the lookup table construction modules, wherein lookup tables are constructed and forwarded to the enhanced luminance image generation module 156 for remapping of the conditioned luminance signal 53.
  • contrast enhancement is provided by heuristic and algorithmic techniques, including the calculation of the image grey-level histogram and sample cumulative distribution function.
  • maximum utilization of available grey-levels for representation of image detail is provided.
  • Entropy is a measure of the information content of the image. Entropy is generally the average number of bits required to represent a pixel of the image. Accordingly, a completely saturated image with only black and white can be represented by one-bit per pixel, on and off, thus, the image entropy is one-bit. A one bit image contains less information than an eight-bit image with 256 shades of grey.
  • optimized entropy is provided while maintaining a monotonic grey-level relationship wherein "monotonic" generally means that the precedence order of the input grey-levels are not changed by the desired grey-level transformation employed.
  • contrast enhancement includes calculation of the sample cumulative distribution function of the image grey-levels.
  • the inverted sample cumulative distribution is applied to map the input grey-levels. Because the inverted sample cumulative distribution function is used, the resulting transform approximates uniformly distributed grey levels within the input range.
  • the uniform density histogram generally has the maximum entropy.
  • the histogram When the image is saturated the histogram has a pathology - an inordinate number of pixels are black or white or both. Direct calculation of the sample cumulative distribution function produces large gaps in endpoints of the grey-level transform. Such gaps mean that the maximum number of grey-level values are not being used, and the resulting output distribution does not closely approximate the desired uniform distribution. Moreover, the resulting image will contain broad areas of a single grey- level (for both black and with saturation). Sometimes, such areas are referred to as "puddles.”
  • FIGURES 8 and 9 are simplified functional block diagrams.
  • the contrast enhancement block diagrams of FIGURES 8 and 9 include: an identity lookup table builder 211, a global equalized lookup table builder 213, a plurality of zone equalized lookup table builders 215, 217, 219 and 221; a plurality of non-linear transform blocks 223, 225, 227, 229 and 231; a global weight functional block 233; a plurality of zone weighting functional blocks 235, 237, 239 and 241; and a plurality of functional operators including adders, multipliers, and dividers.
  • the identity lookup table builder 211 constructs an identity lookup table for the image frame received. Thus, for each pixel having a contrast k, the lookup table output provided by block 211 provides the same contrast, k.
  • the global equalized lookup table builder 213 constructs a global equalized lookup table for the image data received.
  • the functionality of the builder 213 within FIGURE 8 is like that of block 54 within FIGURE 6. Accordingly, a range of useful grey-levels is identified. As described above, the range can include all grey- levels within a cumulative distribution function or any lesser or greater amount as desired, for example all grey levels except for the first and/or last unsaturated grey-level within the distribution function.
  • the range is then scaled to include an extended range of grey-levels including, but not limited to, the full range of grey-levels.
  • the builder 213 provides, as an output, the scaled range as a lookup table to the non-linear transform 223.
  • each of the zone equalized lookup table builders 215, 217, 219 and 221 constructs an equalized lookup table for that portion of the image data within the particular table builder's zone.
  • a range of useful grey-levels is identified by the builder assigned to the zone.
  • the range for each zone can include all grey-levels within a cumulative distribution function, except the first and/or last unsaturated grey-level within the zone's distribution function, or the like.
  • the range is then scaled to include an extended range of grey-levels including, but not limited to, the full range of grey-levels.
  • the lookup table builder for each zone then provides, as an output, its scaled range as a lookup table to a non-linear transform 223, 225, 227, 229 and 231.
  • FIGURE 8 discloses schematically, that the non-linear transforms 223, 225, 227, 229 and 231, provide for weighting grey-level values towards white, thus brightening the image.
  • the non-linear transforms are a power function wherein the input to the non-linear transform for each pixel is normalized over the range of grey level values. For example, if there are 256 grey-level values available for each pixel, then the input for each pixel is normalized for 256 grey-level values. Then, a power is applied wherein the power is preferably a real number in the range of about 0 to about 1 , and preferably 8. Therefore, the output of each non-linear transform 223, 225, 227, 229 and 231 , is a lookup table (no reference number) where the values are skewed towards brightening the image.
  • each non-linear transform 221-231 is multiplied by a weight as shown by blocks 233-241 of FIGURE 8. Accordingly, the weighting provided by global weight 233 determines the balance between combining together two or more inputs. In particular, the global weight determines the balance between the identity lookup table and the transformed global equalized lookup table. The combination or blending together of the identity lookup table 211 and the transformed global equalized lookup table 213 results in a contrast enhanced global lookup table 243. Likewise, the weighting provided by weights 235-241 determines the balance between the contrast enhanced global lookup table 243 and each transformed zone equalized lookup table.
  • each lookup table 245, 247, 249 and 251 receives the original grey- level image data input for each pixel within the image frame.
  • the outputs 263, 265, 267 and 269 of the lookup tables 245, 247, 249 and 251 provide new grey-level values for each pixel within the corresponding zone. This new grey-level values can be increased, decrease, or the same as the original luminance.
  • the outputs 263, 265, 267 and 269, of the lookup tables are multiplied by a corresponding position weighting factor 271, 273, 275 and 277.
  • the new grey-level values for each pixel is multiplied by a weighting factor 271, 273, 275 and 277, based upon the pixel's location relative to one or more reference positions within the image frame.
  • the weighting factors 271, 273, 275 and 277 govern how the grey-level within each zone effects the new grey-level values provided for each pixel within the image.
  • a pixel is effected more by pixel values within the zone within which it is located, and is effected less dramatically by pixel values from a zone furthest from the pixel.
  • the weightings can be normalized so their total sum is 1.
  • the reference positions within the image frame are located at the corners of the image frame. In a further embodiment, the reference positions can be at the centers of the defined zones.
  • the weightings are based upon the distance the pixel is located from the reference positions. For example, if the image frame is a square containing four equally sized zones with the reference positions at the corners of the image frame, then the weighting factors 271, 273, 275 and 277, for the center pixel within the image frame would be 0.25 because the pixel is equal distant from each zone's reference position (i.e., the corners of the image frame). Likewise, each pixel at the corners of the image frame would be weighted such that only the lookup table results for the zone containing the pixel would be applied in determining the pixel's new grey-level. In yet another embodiment, and as will be appreciated by those having ordinary skill in the art, only a single zone can be used within an image frame. Preferably, the zone is centered in the image frame and is small in size than the entire image frame.
  • the device 10 and its combination of elements, and derivatives thereof provide circuits and implement techniques that are highly efficient to implementation yet work very well. Others have proposed various forms of contrast enhancement solutions but the proposals are much more complex to implement. Apparatus and methods according to the invention create an algorithmically efficient procedure that results in cost efficient real-time implementation. This, in turn provides lower product costs.

Abstract

A system and method for manipulating image data is disclosed (Figure 6). Generally, the system and method indentifies (66) useful grey-levels within an input image. The image is then scaled (68) based upon the identified (66) useful grey-levels.

Description

SIGNAL INTENSITY RANGE TRANSFORMATION APPARATUS AND
METHOD
DESCRIPTION
Related Applications
This application claims the benefit of U.S. Provisional Application No. 60/408,663, filed September 6, 2002, and herein incorporated by reference. Technical Field The present invention generally relates to imaging systems, and in particular, to a system for manipulating image data. Background of the Invention
Human interpretation of video or other imagery can be made difficult or even impossible by system noise, image blur, and poor contrast. These limitations are observed, for example, in most video and closed circuit television systems, and others, including such technology as RS-170 monochrome video, NTSC/PAL/SECAM video or digital color video formats.
Extreme lighting variation, for example, due to sunlight beams, can cause typical video cameras and imaging sensors to saturate (i.e., become unable to represent the real- world luminance range) resulting in wide-scale bright and dark regions having extremely low-contrast wherein objects are difficult or impossible to discern. At outdoor automated teller machines, for example, sunlight beams or strong lighting, in the background can cause a person in a dark area to become unidentifiable due to low contrast. This weakness is due to the limited luminance range of the imaging system. The electronic iris and automatic gain control provided by some imaging systems are designed to try to optimally map the wide luminance values within a real- world light situation into a limited range digital representation, often resulting in a poor compromise. To adequately represent the bright areas, less illuminated areas become dramatically compressed in contrast and thus become very dark.
Besides having limited range, video or imagery from lower-cost imaging sensors can have significant noise due to a number of basic system limitations, as well as significant blur due to lower-cost, small, or simple optical configurations. Reducing noise and blur within these systems can improve the ability for a human viewer to effectively interpret image content.
Moreover, digital samples of interlaced analog video from a video field are typically taken by imaging systems. The noise inherent in such digital samples can make human interpretation of important details in the image difficult.
Hence, a need exists for a luminance range transformation apparatus and method that manipulates image data for improved interpretation thereof.
Others have provided some image post processing devices which can enhance contrast of an image. However, in many areas such as in security monitoring, real-time evaluation of images is highly beneficial or necessary. Accordingly, there is also a need to provide a luminance range transformation apparatus which can enhance an image in real-time or near-real time. Summary of the Invention
According to the present invention, a system and method has been developed that considers the mechanisms and defects of imagery from video or other sources to prescribe a sequence of tailored image processing operations designed to improve human interpretation of resulting images.
However, in a broad aspect of the invention, methods and devices are provided to redistribute discrete signal intensity values within groups of signal intensity values. The signal intensity values are supplied directly or indirectly from a sensor sensing an environment. It is proposed that the inventions can be used for enhanced interpretation of any array of signal intensities or variable values in any group of such values that have spatial or geometric relationships to one another (e.g. coordinates). For example, the detailed disclosures below are directed to redistribution of grey-scale or luminance values in a rectilinear array (pixels) from a camera. It is also contemplated however, that the inventions may be applied to other values in a video image, such as redistribution of chrominance values. It is also proposed that the inventions may be employed with other sensing modalities such as, magnetic resonance imagery, radar, sonar, infrared, ultraviolet, microwave, X-ray, radio wave, and the like. According to another aspect of the invention, the spatial scale of an entire group of signal intensity values are considered, for example, the luminance in an entire pixel image, so that the overall brightness from corner to corner is taken into account. In an orthogonal direction, all the frequency content of an image must be represented by the one global mapping, including the very lowest spatial frequency. Often, this single global mapping has to stretch to map all of the low frequency variation in the image, and then fails to enhance higher frequency structure of interest. The result can be a bright "bloom" on one side of the image, with too dark an area on the other side of the image. As such, there may not be optimal recognition of spatial structure because of the need to represent the large scale variation across the entire image.
It is proposed that to further improve overall contrast balance to reveal image structure at scales of interest by applying equalization at spatial scales representative of scales of interest. Accordingly, the inventions propose generating subset partitions of an image (group of signal intensities) representative of the spatial scales of interest. A transform mapping (e.g. of luminance) for the subsets of signal values is generated, so that a spatial scale (e.g., is 1/4 of a global image for a quadrant subset) so as to mitigate or eliminate the lowest frequency from consideration. This improves representation of contrast for structures at this scale. Computing a luminance transformation at every pixel (with for example a filter kernel) for a neighborhood around that pixel, would result in a large computational burden because the resulting spatial scales are too small. In contrast, the present invention employs accumulating pixel samples across larger "right-sized" spatial scales, while using efficient interpolation to produce the correct transform (e.g. for luminance) representation at each pixel.
According to another aspect of the invention, apparatus and methods include: decomposing a group of signal values into subgroup partitions from which to sample values and construct associated transforms and to combine those transform values at every coordinate (e.g. pixel) according to a rule which weights the contribution of each mapping in accordance with "image" geometry and value coordinate or position.
According to another aspect of the invention, apparatus and methods are provided for blending transformed values from the global set of values (e.g. entire pixel image) with transformed values from the spatial segments to adjust contributions from several spatial scales as desired.
While overall interpretation of groups of signal intensity values is provided according to the invention, it is of particular note that the effectively mitigate signal values which are out of limit for a sensor system, for example, saturated sensor response. In a preferred embodiment for enhancing luminance contrast in video signals, preferred operations can include digital sampling and luminance/color separation, noise reduction, deblurring, pixel noise reduction, histogram smoothing, contrast stretching, and luminance and chrominance re-combination. One or more of the operators can have configurable attributes, such as degree of noise reduction, brightness, degree of deblurring, and determined range of useful grey-levels.
Other advantages and features of the present invention will be apparent from the following description of a specific embodiment illustrated in the accompanying drawings. Brief Description of Drawings
FIGURE 1 is a simplified block diagram of a device in accordance with the present invention, including a logical device;
FIGURE 2 is a simplified functional block diagram of a process performed by the logical device of FIGURE 1 , the process having a color transform, an image pre-contrast conditioner, a contrast enhancement, an image post-contrast conditioner, and a color inverse transform;
FIGURE 3 is a simplified functional block diagram of the image pre-contrast conditioner of FIGURE 2, the pre-contrast conditioner comprising a system noise reduction filter, a deblurring module, and a pixel noise reduction module;
FIGURES 4(a)-(f) depict various exemplary kernels shapes that can be used with the system noise reduction filter, deblurring module, and pixel noise reduction module of FIGURE 3;
FIGURE 5 is a simplified functional block diagram of the contrast enhancement block of FIGURE 2, the contrast enhancement comprising an equalized lookup table construction block and an enhanced luminance image generation block;
FIGURE 6 is a simplified functional block diagram of the equalized lookup table construction block of FIGURE 5;
FIGURE 7 is a simplified functional block diagram of another embodiment of the contrast enhancement block of FIGURE 2; and,
FIGURES 8 and 9 is a simplified function block diagram of yet another embodiment of the contrast enhancement block of FIGURE 2. Detailed Description
This invention is susceptible of embodiments in many different forms. For example, the methods and apparatus disclosed there is shown in the drawings and will herein be described in detail, a preferred embodiment of the invention. The present disclosure is to be considered as an exemplification of the principles of the invention and is not intended to limit the broad aspect of the invention to the embodiment illustrated. Referring now to the drawings, and as will be appreciated by those having skill in the art, each of the FIGURES depicts a simplified block diagram wherein each block provides hardware (i.e., circuitry), firmware, software, or any combination thereof that performs one or more operations. Each block can be self-contained or integral with other hardware, firmware, or software associated with one or more other blocks.
Turning particularly to FIGURE 1, a device 10 is disclosed for enhancing, through transformation, the luminance range of an image input signal. The device 10 includes an input connector 12, logic circuitry 14, and an output connector 16. The connectors 12 and 16 are mounted in a conventional manner to an enclosed housing 13, constructed of a metal, metal alloy, rigid plastic, or combinations of the above, that contains the logic circuitry 14. In one embodiment, one or more of the modules described herein are performed by the logic circuitry 14 comprising of one or more integrated circuits, commonly referred to as "ICs," placed on one or more printed circuit boards mounted within the housing 13.
Preferably, the device 10 is a stand-alone or embedded system. As used herein, the term "stand-alone" refers to a device that is self-contained, one that does not require any other devices to perform its primary functions. For example, a fax machine is a stand-alone device because it does not require a computer, printer, modem, or other device. Accordingly, in an embodiment, the device 10 does not need to provide ports for connecting a disk drive, display screen, or a keyboard. However, in an alternative embodiment, the device 10 could provide one or more ports (e.g., RS-232) for supporting field interactivity.
Also, as used herein, an embedded system is a system that is not a desktop computer or a workstation computer or a mainframe computer designed to admit facile human interactivity. Another delineator between embedded and "desktop" systems is that desktop systems (and workstation, etc.) present the status of the computer state to the human operator via a display screen and the internal state of the computer is represented by icons on the screen, and thus the person can interact with the computer internal state via control of the icons. Moreover, such a computer uses a software layer called an "operating system" through which the human operator can interact with the internal state of the computer. Conversely, with an embedded system, while it is performing its work function, the human operator cannot interact with the work process except to stop it.
The input connector 12 provides for operably connecting the device 10 to an image input signal 17 generated by a video camera (not shown), or the like, having a video output. In one embodiment, the input connector 12 consists of an F connector, BNC connector, RCA jacks, or the like. The input connector 12 is operably connected to the logic circuitry 14 by way of a conductive path attached to the input connector and the printed circuit board contained within the housing 13. The logic circuitry could also be coupled through other than a conductive path such as through optical coupling.
Preferably, but not necessarily, the image input signal 17 is a conventional analog video signal containing a plurality of still images or fixed image frames taken in a sequential manner. Each frame provided by the image input signal is also referred to herein as an image input frame. Each image or frame includes data regarding an array of pixels contained therein.
The output connector 16 of the device 10 provides for connecting the device to an output device such as a monitor (not shown). Like the input connector 12, the output connector 16 consists of any means for outputting the signal to other devices such as, an F connector, BNC connector, RCA jacks, or the like. The output connector 16 is operably connected to the logic circuitry 14 by way of a conductive or coupled path attached to the output connector and the printed circuit board contained within the housing 13. As explained in detail further herein, the output signal provided by connector 16, and thus the device 10, provides an output signal which includes data resulting from transforming or other operations carried out with respect to image input signal 17 received ("transformed output signal"). The output signal can include a plurality of image output frames and be formatted as a conventional analog video signal, a digital signal, or the like. For example, but by no means exclusive, the output signal can be in a format as defined by NTSC, VGA, HDTV, or other desired output formats. In one embodiment, the logic circuitry 14 within the device 10 includes, inter alia, circuitry configured to transform the variable range of grey-scale values in the image input signal 17 received by the input connector 12. Preferably, the logic circuitry 14 includes a logical device 18 with corresponding support circuitry 20, a video decoder 22, and a video encoder 23. The support circuitry 20 preferably includes a microcontroller 24, a read only memory 26, and a random access memory 28 comprising a synchronous dynamic random access memory.
In an embodiment, an optional switch 29 is provided for configuring the logic circuitry 14 within the device 10. The switch 29 is operably connected to the microcontroller 24 and logical device 18. The switch 29 allows a user to enable or disable, features or processes provided by the logic circuitry 14 within the device 10. In one embodiment, the switch 29 consists of a conventional DIP switch. In an alternative embodiment, the configuration of the circuitry 14 within the device 10 is hardwired, or can be set via software commands, instead of using a switch.
Preferably, video decoder 22 is operably connected to the input connector 12 and the logical device 18. Accordingly, the video decoder 22 receives the image input signal 17 that can consist of live video from a television broadcast, a video tape, a camera, or any other desired signals containing or representing image content. The video decoder 22 preferably is a conventional device for tracking the video image input signal 17, digitizing the input signal (if required), separating out the brightness and color information from the input signal, and forwarding the digital video signal 30 to the logical device 18 on a frame by frame basis.
In one embodiment, the input image signal 17 received by the video decoder 22 is an analog signal formatted in a predefined manner such as PAL, NTSC, or another conventional format. The video decoder 22 converts the analog signal 17 into a digital video signal 30 using a conventional analog-to-digital conversion algorithm . Preferably, the digital video signal 30 provided by the video decoder 22 includes luminance information and color information in any conventional manner such as, specified by YUV format, YCbCr format, super video, S-video, or the like. Alternatively, the digital video signal 30 can have the luminance information embedded therein such as that provided by digital RGB, for example.
In one embodiment, the video decoder 22 is capable of converting a plurality of different analog video formats into digital video signals suitable for processing by the logical device 18 as described in detail further herein. In one embodiment, the microcontroller 24 configures the video decoder 22 for converting the image input signal
17 into a digital video signal 30 having a specific format type (e.g., CCIR601, RGB, etc.). If desired, the microcontroller 24 determines the format of the image input signal 17, and configures the video decoder 22 accordingly. The determination can be accomplished by the microcontroller 24 checking the user or device manufacturer configured settings of the DIP switch corresponding with the format of the image input signal 17 expected to be received. Alternatively, the video decoder 22 can include circuitry for automatically detecting the format of the image input signal 17, instead of using preset DIP switch settings.
Preferably, the gain of the video decoder 22 is set to reduce overall contrast on the luminance for reducing the probability of image saturation. In an alternative embodiment, the video decoder 22 provides a fixed bit resolution output range (e.g., 8 bit, 16 bit, 32 bit, etc.) and a digitizer maps the image input signal in a conventional manner for effective use of the resolution output range. Preferably, the full range of the digitizer output is utilized. Turning to the logical device 18, as will be understood by those having skill in the art, the logical device receives digital video signals 30 and provides digital output signals 31 (i.e., transformed pixel array or frame data) in response to the digital video signals 30. As indicated previously, in one embodiment, the device 10 receives analog image input signals 17 that are converted by the video decoder 22 into digital video signals 30 for manipulation by the logical device 18. Moreover, as explained in detail further herein, the digital output 31 of the logical device 18 can be converted (if desired) into an analog output by the video encoder 23 connected between the logical device 18 and the output connector 16.
Preferably, the logical device 18 consists of a conventional field programmable gate array (FPGA). However, the logical device 18 can consists of a Digital Signal Processor (DSP), a microcontroller, an Application Specific Integrated Circuit (ASIC), or other suitable device. In an alternative embodiment, instead of receiving analog input signals 17, the device 10 can receive digital image input signals that are provided, without any conversion, to the logical device 18. Thus, in this alternative embodiment, the video decoder 22 can be omitted.
Regardless of whether a video encoder or decoder is provided, it is to be understood that the logical device 18 is operably connected between connectors 12 and
16. Accordingly, an image input signal enters the input connector 12, is modified by the logical device 18, and then exits via the output connector 16 as a transformed output signal. Preferably, the frame output rate of the device 10 is substantially equal to the frame input rate to the device. The device 10 provides an output of thirty frames per second in response to a frame input rate of thirty frames per second.
In one embodiment, the logical device 18 provides a sequence of image processing functional modules or blocks 32 within a process as illustrated in FIGURE 2. In one embodiment, the blocks 32 represent a color transform 34, an image pre- contrast conditioner 36, a contrast enhancement 38, an image post-contrast conditioner 40, and a color inverse transform 42.
Each of the modules 32 preferably performs a specific functional step or plurality of functional steps as described in detail further herein. Turning back to FIGURE 1, static data for configuring the logical device 18, if needed or desired, is stored within the read only memory 26 operably connected thereto. As appreciated by those having skill in the art, the read only memory 26 provides for storing data that is not alterable by computer instructions.
Storage for data manipulation and the like is provided by the synchronous dynamic random access memory (SDRAM) 28. Preferably, a high speed random access memory is provided by the support circuitry 20 to reduce performance bottlenecks. In one embodiment, the memory 20 is used as a field buffer.
Turning back to FIGURE 2, the color transform 34 provides for separating the luminance from the digital video signal 30, as desired. In one embodiment, the digital video signal 30 provided by the video decoder 22 consists of a digital RGB signal. As such, luminance information is not separated from color information. Thus, the color transform 34 provides for separating the luminance information for each pixel, within each frame, from the digital video signal 30.
Preferably, the color transform 34 uses a conventional algorithm for converting a digital RGB video input signal into a Yuv format signal, YCbCr format signal, or other desired format signal. Accordingly, the color transform 34 provides an output comprising three channels: luminance 43, and U-V or other values ascribed per pixel location.
In an alternative embodiment, the digital video signal 30 provided by the video decoder 22 consists of a super video or S-video. As such, the digital video signal 30 consists of two different signals: chrominance and luminance. Accordingly, the color transform 34 can be omitted from the process of FIGURE 2 because the luminance 43 is separately provided by the S-video input.
It should be understood as noted above, that a separate channel or channels (not shown) comprising the steps to be discussed below with respect to luminance information, could be provided to any one of RGB/chrominance values independently for transforming the values and redistributing the color values in the image. As shown in FIGURE 2, the luminance information 43 contained within the digital video signal 30 is received by the image pre-contrast conditioner 36. As shown in FIGURE 3, the image pre-contrast conditioner 36 can include a video time integration module or block 44, a system noise reduction module or block 46, a deblurring module or block 48, and a pixel noise reduction module or block 50. As will be appreciated by those having ordinary skill in the art, any one or all of the blocks within FIGURE 3 can be omitted including, but not limited to, the video time integration module 44.
In one embodiment, the video time integration module 44 consists of a low-pass filter for noise reduction of the luminance signal 43. As understood by those having skill in the art, each frame includes an array of pixels wherein each pixel has a specific i,j coordinate or address. Noise reduction by the video time integration module 44 is preferably achieved by combining the current frame (n) with the immediately preceding frame (n-1) stored by the logical device 18 within the memory 28. Desirably, for each pixel c(ij) in the current image frame (n), the corresponding pixel p(ij) from the preceding captured frame (n- 1 ) is subtracted to result in a difference d(i,j). The absolute value of the difference d(ij) is determined, and compared to a predetermined threshold value. If the absolute value of the difference d(i,j) is greater than the predetermined threshold, then the time integration output signal 45 of the video time integration module 44 is c(i j). Otherwise, the time integration output signal 45 of the video time integration module 44 is the average of c(ij) and p(i,j).
As indicated above, the video time integration module 44 can operate such that object motion, which ordinarily creates large pixel differences, are represented by current pixels while background differences due to noise are suppressed by the averaging. Accordingly, the video time integration module 44 provides for avoiding image blurring as a result of low-pass spatial domain filtering on the current image.
In one embodiment, the median filter or system noise reduction module 46 receives the time integration output signal 45 (i.e., transformed pixel frame data) and, in response thereto, provides for reducing pixel noise caused by the system senor (e.g., camera), electromagnetic, and/or thermal noise. The median filter 46 can include a user selectable level of noise reduction by selecting from a plurality of filter kernels via DIP switch 29, or the like, operably coupled to the logical device 18. The filter kernel applied by the noise reduction module 40 to the current frame, as preferably modified by time integration module 44, is designed to achieve noise reduction while minimizing the adverse effect of the filter on perceived image quality. For instance, a 3X1 kernel (FIGURE 4(a)) provides for a reduction in row noise due to the kernel shape, while having minimal adverse effect on the image. Also, a 5-point or plus kernel (FIGURE 4(b)) offers symmetric noise reduction with low adverse impact to image quality. Moreover, a hybrid median filter (FIGURE 4(c)) offers symmetric noise reduction with lower adverse impact than the full 3X3 kernel. In one embodiment, a user or the manufacturer can select from the kernel shapes shown in FIGURES 4(a)-(c).
The deblurring module or block 48 provides for counteracting image blurring attributable to the point spread function of the imaging system (not shown). In an embodiment, the deblurring module or block 48 uses a Laplacian filter to sharpen the output 47 (i.e., transformed pixel frame data) received from the filter 46. The deblurring module or block 48 can include a user-selectable level of image sharpening from a plurality of Laplacian filter center pixel weights. In one embodiment, the user or manufacturer can select the center pixel weight via the DIP switch 29 operably coupled to the logical device 18.
Preferably, the Laplacian filter uses a 3X3 kernel (FIGURE 4(d)) with selectable center pixel weights of 9, 10, 12, or 16, requiring normalization divisors of 1, 2, 4, or 8, respectively. As shown in FIGURE 4(d), the remaining kernel pixels are preferably assigned a weight of -1. The result of applying the kernel is normalized by dividing the convolution result for the current pixel element by the normalization divisor.
As will be appreciated by those having ordinary skill in the art, the deblurring module or block 48 can use other filters in place of a Laplacian type filter. In an embodiment, and in like fashion to the use of the Laplacian filter with various weights, a general filter with kernel as shown in FIGURE 4(e) can be used for deblurring.
The output 49 (i.e., transformed pixel frame data) of the deblurring module 48 emphasizes isolated pixel noise present in the input image signal. Preferably, the pixel noise reduction module 50 provides for reducing isolated pixel noise emphasized within the output 49 of the deblurring module 48. In one embodiment, the pixel noise reduction module 50 employs a small-kernel median filter (i.e., small pixel neighborhood such as a 3X3) to reduce isolated pixel noise. The pixel noise reduction filter or module 50 can provide a user or manufacturer selectable level of noise reduction by allowing for the selection from a plurality of filter kernels. In a preferred embodiment, the DIP switch
29, software, or the like, is used to select from a five-point "plus patterned" kernel (FIGURE 4(b)) or a hybrid median filter such as (FIGURE 4(c)).
As shown in FIGURE 2, the contrast enhancement module or block 38 enhances the contrast of the conditioned luminance signal 53 (i.e., transformed pixel frame data) provided by the image pre-contrast conditioner 36 in response to the original luminance input signal 17. In an alternative embodiment, no image pre-contrast is provided. Instead, the contrast enhancement module 38 directly receives the image input signal 17, and in response thereto, generates an enhanced image signal 55.
Turning to FIGURE 5, the contrast enhancement module or block 38 preferably includes a sample area selection module or block 52, an equalized lookup table construction module or block 54, and an enhanced luminance generation module or block 56. As used herein, and as will be appreciated by those having ordinary skill in the art after reading this specification, the phrases "lookup table" and "equalized lookup table" both refer to a lookup table derived from grey-level values or a lookup table representative of the inversion of the cumulative distribution function derived from an accumulated histogram.
In one embodiment, the sample area selection module or block 52 receives the conditioned luminance image signal 53 provided by the image pre-contrast conditioner 36 (FIGURES 2 and 3) and a selected portion of the input image, via input 57, to be enhanced. In response to the conditioned luminance input signal 53 and the image portion selection 57, the sample area selection module provides a selected conditioned luminance image signal 59 comprising conditioned luminance pixel data from the pixels within the selected region.
Depending upon the overall embodiment of a system, the selected portion of the image can be selected in a conventional manner such as by using a point-and-click interface (e.g., a mouse) to select a rectangular sub-image of an image displayed on a monitor (not shown). Alternatively, the selected portion of the image can be selected by automated means such as by a computer, based upon movement detection, with a portion of the image or other criteria.
In one embodiment, the equalized look-up table construction module or block 54 receives the selected conditioned image signal 59 from the sample area selection module or block 52 and, in response thereto, provides for creating a lookup table 61 that is used in generating the enhanced image signal 55 (i.e., transformed pixel frame data).
As previously indicated, the selected conditioned luminance signal 59 received from the sample area selection module or block 52 can be a sub-image comprising a portion of the overall conditioned luminance image signal 53 received from the image pre-contrast conditioner 36 (FIGURES 2 and 3). Alternatively, the selected conditioned luminance signal 59 can consist of the complete conditioned luminance signal 53, without any apportioning by the sample area selection module 52.
As shown in FIGURE 6, the equalized lookup table construction module, or block 54 of FIGURE 5, includes an input smoothing module or block 58, a histogram accumulation module or block 60, a histogram smoothing module or block 62, a cumulative distribution function integration module or block 64, a saturation bias removal module or block 66, a linear cumulative distribution function scaling module or block 68, and a lookup table construction module or block 70. In one embodiment, the input image smoothing module, or block 58, receives the selected conditioned luminance image signal 59 and, in response thereto, generates a smoothed conditioned luminance image signal on a frame-by-frame basis. Preferably, the input image smoothing module or block 58 applies a Gaussian filter with a traditional 3X3 kernel to the selected signal 59 for smoothing the image within each frame. As recognized by those having skill in the art, the Gaussian filter performs a weighted sum (center pixel = highest weight), wherein the result is normalized by the total kernel weight.
In one embodiment, a histogram accumulation module, or block 60, provides for generating a histogram of the selected conditioned image signal 59, as modified by the input image smoothing module. The histogram accumulation module, or block 60, accumulates a conventional histogram from the received input. Accordingly, the histogram can be used to provide a graph of luminance levels to the number of pixels at each luminance level. Stated another way, the histogram reflects the relationship between luminance levels and the number of pixels at each luminance level.
Accordingly, the histogram accumulation module or block 60 includes a plurality of data storage locations or "bins" for tracking the number of occurrences of each grey-level occurring in the received input signal, for example, such as that received by block 58. Preferably, the histogram accumulation module or block 60 includes a bin for every discrete grey-level in the input signal. As stated previously, for each grey-level in the input signal, the histogram accumulation module or block 60 determines the number of pixels in the input luminance image with the corresponding grey-level after being filtered, if desired. Accordingly, the histogram result for any given grey-level k is the number of pixels in the luminance image input having that grey-level. In one embodiment, the histogram smoothing module or block 62 provides for reducing noise in the histogram created from the input image. The histogram smoothing module, or block 62, receives the histogram from the histogram accumulation, module or block 60, and in response thereto, generates a smoothed histogram. Preferably, the histogram smoothing module or block 62 applies a 5-point symmetric kernel (FIGURE 4(f)) to filter the histogram calculated for each frame.
In one embodiment, the cumulative distribution function integration module or block 64 provides for integrating the histogram to create a cumulative distribution function. The cumulative distribution function integration module or block 64 receives the histogram from the histogram smoothing module 62, or optionally from the histogram accumulation module 60. The cumulative distribution function integration module 64 integrates the received histogram to generate a cumulative distribution function. It has been observed that integrating a smoothed histogram can result in a cumulative distribution function having an increased accuracy over a cumulative distribution function resulting from an unsmoothed histogram.
The cumulative distribution function integration module or block 64 includes a plurality of data storage locations or "bins" which hold the cumulative distribution function result for each input grey- level. Preferably, the cumulative distribution function integration module, or block 64, includes a bin for every discrete grey-level in the image resolution input range. For each grey-level in the image resolution input range, the cumulative distribution function integration module or block 64 determines the cumulative distribution function result and stores that result in the bin corresponding to that grey-level. For each grey-level k, the cumulative distribution function result is the sum of the histogram value for grey-level k and the cumulative distribution function result for the previous grey-level k- 1. Accordingly, the following equation describes the cumulative distribution function result for a given grey- level k:
CDF(k) = H(k) + CDF(k-l) where CDF(k) is the cumulative distribution function result for grey-level k, H(k) is the histogram value for grey-level k, and CDF(k-l) is the cumulative distribution function result for the grey-level one increment lower than grey-level k.
The saturation bias identification module, or block 66, provides for identifying the grey-levels that bound the useful information in the luminance signal 43. In one embodiment, the saturation bias identification module 66 receives the cumulative distribution function from the cumulative distribution function integration module 64 and determines the grey-levels at which the first unsaturated grey-level, kf, and the last unsaturated grey-level, k,, occurs. The first unsaturated grey-level, kf, is determined by identifying the first grey-level ko for which the cumulative distribution function returns a non-zero value. The grey-level ko is treated as saturated, and the saturation bias identification module 66 identifies kf as the sum of ko plus one additional grey-level. The last unsaturated grey-level, kb is determined by identifying the first grey-level k„ for which the cumulative distribution function returns the number of pixels in the image. The grey-level kn is treated as saturated, and the saturation bias identification module or block 66 identifies k, as the difference between k„ minus one additional grey-level.
Accordingly, the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, kg plus one additional grey- level through kn minus one additional grey-level. In another embodiment, the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, ko through kn minus one additional grey-level. In yet another embodiment, the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including ko plus one additional grey-level through kn. In a further embodiment, the useful grey-levels identified by the saturation bias identification module 66 is the range from, and including, ko plus X additional grey-level(s) through kn minus Y additional grey-level(s), wherein X and Y are whole numbers greater than zero.
The linear cumulative distribution function scaling module or block 68 provides for scaling the portion of the cumulative distribution function corresponding to the unsaturated, useful grey-levels in the luminance input image across the entire range of cumulative distribution function grey-level inputs. In a preferred embodiment, the linear cumulative distribution function scaling module 68 receives the cumulative distribution function, provided by the cumulative distribution function integration module 64, and the first unsaturated grey-level kf and the last unsaturated grey-level k, from the saturation bias identification module 66.
The linear cumulative distribution function scaling module 68 includes a plurality of data storage locations or "bins" equal to the number of bins in the cumulative distribution function for holding the linearly mapped cumulative distribution function result for each input grey-level. In a preferred embodiment, the cumulative distribution function scaling module 68 scales the portion of the cumulative distribution function between kf and k, inclusively across the entire range of the linearly scaled cumulative distribution function. Accordingly, for each grey-level k in the input range, the linearly scaled cumulative distribution function output value LCDF(k) is determined by the following equation:
LCDF(k) =(CDF(k) - CDF(kf))/(CDF(Kl) - CDF(Kf))
where CDF(k) is the cumulative distribution function result for grey-level k, CDF(kf) is the cumulative distribution function result for kf, and CDF(kl) is the cumulative distribution function result for kl. Each linearly mapped cumulative distribution function result is stored in the bin corresponding to the current grey-level k. If LCDF(k) is negative for any input grey-level k, the linearly scaled cumulative distribution function result in the bin corresponding to the input grey-level k is set to zero.
The determination of the cumulative distribution function output value can be calculated using known methodologies for improving computation ease. For instance, the numerator can be scaled by a multiplier, before the division operation, to obtain proper integer representation. The result of the division can then be inversely scaled by the same multiplier to produce the LCDF(k) value. Preferably, the scale factor is an adequately large constant, such as the number of pixels in the input image.
In one embodiment, the lookup table construction module or block 70 generates a lookup table 61 that is used by the enhanced luminance image generation module (FIGURE 5) to produce the enhanced luminance image signal 55. The lookup table construction module 70 receives the linearly scaled cumulative distribution function from the linear cumulative distribution function scaling module 70. In response to the scaled cumulative distribution function, the lookup table construction module 70 provides the lookup table 61.
Preferably, the lookup table 61 is constructed by multiplying each linearly scaled cumulative distribution function result by the number of discrete grey-levels in the output image resolution range which can be provided by the DIP switch 29 (FIGURE 1), software, or other desired means. If the multiplication result is greater than the maximum value in the output image resolution range for any input grey-level, the linearly scaled cumulative distribution function result for that input grey-level is set to the maximum value in the output image resolution range. Accordingly, for each grey- level input in the linearly scaled cumulative distribution function, the result is a value in the output image resolution range. Turning back to FIGURE 5, the enhanced luminance image generation module, or block 56, is responsive to the lookup table 61 and the conditioned luminance signal 53. The luminance of each pixel within each input frame can be reassigned a different value based upon the lookup table corresponding to the same frame. Stated another way, the luminance value for each pixel in the conditioned luminance input image is used as an index in the look-up table, and the value at that index location is used as the luminance value for the corresponding pixel in the enhanced luminance output image. The reassigned frame data is then forwarded, preferably to image post contrast conditioner 40, as the enhanced image signal 55.
As shown in FIGURE 2, the image post contrast conditioner 40 provides for filtering the enhanced image signal 55 to generate an enhanced conditioned image signal
72. The conditioner 40 can apply to the image signal 55 one or more conventional imaging filters, including an averaging filter, a Gaussian filter, a median filter, a Laplace filter, a sharpen filter, a Sobel filter, and a Prewitt filter. The color inverse transform 42 generates the contrast enhanced digital video output signal 31 , having an expanded range, in response to the enhanced image signal 72 and the color information for the same frame. The enhanced image signal and color information are combined in a conventional manner wherein luminance is combined with color information.
As will be appreciated by those having skill in the art, while the luminance of the input signal is being enhanced, the color information can be stored within memory 28, for example. Once the luminance has been enhanced, it can be re-combined with the color information stored in memory. In response to the digital video output signal 31, the video encoder 23 preferably provides an analog output signal via output connector 16. The video encoder 23 preferably utilizes conventional circuitry for converting digital video signals into corresponding analog video signals.
Turning to FIGURE 7, an alternate embodiment of the contrast enhancement module is depicted. Within FIGURE 7, that last two digits of the reference numbers used therein correspond to like two digit reference numbers used for like elements in
FIGURES 5 and 6. Accordingly, no further description of these elements is provided.
The median bin identification module or block 174 determines the median of the histogram provided by the histogram accumulation module 160. Next, the brightness midpoint value is calculated by the weighted midpoint calculation module or block 176.
The brightness midpoint value preferably is calculated as the weighted sum of the actual grey-level midpoint of the input frame and the median of the input distribution. Preferably, the weights assigned to the grey-level midpoint and the median of the input distribution are configured as compile-time parameters. The image segmentation module, or block 178, receives the luminance image data for the selected sample area, via module 152, and provides for segmenting the luminance image data into light and dark regimes and performing enhancement operations separately (i.e., luminance redistribution) on each of the regimes. Stated another way, the image segmentation module 178 assigns each pixel in the selected sample area to either the light or dark regime based on each pixel's relationship to the brightness midpoint value received from the midpoint calculation module 176. Pixels with a brightness value greater than the midpoint are assigned to the light regime, and pixels with a brightness value less than the midpoint value are assigned to the dark regime.
Pixel values assigned by the image segmentation module 178 are received by the lookup table construction modules, wherein lookup tables are constructed and forwarded to the enhanced luminance image generation module 156 for remapping of the conditioned luminance signal 53.
According to broad aspects of the invention, contrast enhancement is provided by heuristic and algorithmic techniques, including the calculation of the image grey-level histogram and sample cumulative distribution function. Preferably, maximum utilization of available grey-levels for representation of image detail is provided. Entropy is a measure of the information content of the image. Entropy is generally the average number of bits required to represent a pixel of the image. Accordingly, a completely saturated image with only black and white can be represented by one-bit per pixel, on and off, thus, the image entropy is one-bit. A one bit image contains less information than an eight-bit image with 256 shades of grey. According to broad aspects of the invention, optimized entropy is provided while maintaining a monotonic grey-level relationship wherein "monotonic" generally means that the precedence order of the input grey-levels are not changed by the desired grey-level transformation employed.
In one embodiment, contrast enhancement includes calculation of the sample cumulative distribution function of the image grey-levels. The inverted sample cumulative distribution is applied to map the input grey-levels. Because the inverted sample cumulative distribution function is used, the resulting transform approximates uniformly distributed grey levels within the input range. The uniform density histogram generally has the maximum entropy.
As known by those having skill in the art, classical "histogram equalization" directly applies the inverted sample cumulative distribution function to form the grey- level mapping lookup table. In an embodiment in accordance with the present invention, along with the inverted sample cumulative distribution data, it is preferred to combine a linear mapping to ensure full-range output, and also combine a saturation bias identification to improve grey-level uniformity in the presence of saturation. As will be appreciated by those having skill in the art, image saturation creates a pathology in the inverted sample cumulative distribution mapping that is not addressed in classical histogram equalization.
When the image is saturated the histogram has a pathology - an inordinate number of pixels are black or white or both. Direct calculation of the sample cumulative distribution function produces large gaps in endpoints of the grey-level transform. Such gaps mean that the maximum number of grey-level values are not being used, and the resulting output distribution does not closely approximate the desired uniform distribution. Moreover, the resulting image will contain broad areas of a single grey- level (for both black and with saturation). Sometimes, such areas are referred to as "puddles."
An embodiment of another broad aspect of the invention is disclosed in FIGURES 8 and 9 which are simplified functional block diagrams. The contrast enhancement block diagrams of FIGURES 8 and 9 include: an identity lookup table builder 211, a global equalized lookup table builder 213, a plurality of zone equalized lookup table builders 215, 217, 219 and 221; a plurality of non-linear transform blocks 223, 225, 227, 229 and 231; a global weight functional block 233; a plurality of zone weighting functional blocks 235, 237, 239 and 241; and a plurality of functional operators including adders, multipliers, and dividers.
The identity lookup table builder 211 constructs an identity lookup table for the image frame received. Thus, for each pixel having a contrast k, the lookup table output provided by block 211 provides the same contrast, k.
The global equalized lookup table builder 213 constructs a global equalized lookup table for the image data received. In particular, the functionality of the builder 213 within FIGURE 8 is like that of block 54 within FIGURE 6. Accordingly, a range of useful grey-levels is identified. As described above, the range can include all grey- levels within a cumulative distribution function or any lesser or greater amount as desired, for example all grey levels except for the first and/or last unsaturated grey-level within the distribution function. The range is then scaled to include an extended range of grey-levels including, but not limited to, the full range of grey-levels. The builder 213 provides, as an output, the scaled range as a lookup table to the non-linear transform 223.
Similar to the global equalized lookup table builder 213, each of the zone equalized lookup table builders 215, 217, 219 and 221, constructs an equalized lookup table for that portion of the image data within the particular table builder's zone. In particular, within each zone, a range of useful grey-levels is identified by the builder assigned to the zone. The range for each zone can include all grey-levels within a cumulative distribution function, except the first and/or last unsaturated grey-level within the zone's distribution function, or the like. The range is then scaled to include an extended range of grey-levels including, but not limited to, the full range of grey-levels. The lookup table builder for each zone then provides, as an output, its scaled range as a lookup table to a non-linear transform 223, 225, 227, 229 and 231.
FIGURE 8 discloses schematically, that the non-linear transforms 223, 225, 227, 229 and 231, provide for weighting grey-level values towards white, thus brightening the image. In an embodiment, the non-linear transforms are a power function wherein the input to the non-linear transform for each pixel is normalized over the range of grey level values. For example, if there are 256 grey-level values available for each pixel, then the input for each pixel is normalized for 256 grey-level values. Then, a power is applied wherein the power is preferably a real number in the range of about 0 to about 1 , and preferably 8. Therefore, the output of each non-linear transform 223, 225, 227, 229 and 231 , is a lookup table (no reference number) where the values are skewed towards brightening the image.
The output of each non-linear transform 221-231 is multiplied by a weight as shown by blocks 233-241 of FIGURE 8. Accordingly, the weighting provided by global weight 233 determines the balance between combining together two or more inputs. In particular, the global weight determines the balance between the identity lookup table and the transformed global equalized lookup table. The combination or blending together of the identity lookup table 211 and the transformed global equalized lookup table 213 results in a contrast enhanced global lookup table 243. Likewise, the weighting provided by weights 235-241 determines the balance between the contrast enhanced global lookup table 243 and each transformed zone equalized lookup table. These combinations or blends result in a plurality of zonal balanced lookup tables 245-251 wherein the balanced lookup tables are contrast enhanced and relate, in part, to a particular zone of the original input image. Turning to FIGURE 9, the outputs from the zonal balanced lookup tables 245 ,
247, 249 and 251 of FIGURE 8, are combined with corresponding pixel position weights for determining relative zonal pixel grey-levels 253, 255, 257 and 259. The relative zonal pixel grey-levels 253, 255, 257 and 259, are summed together to provide a final or composite enhanced pixel grey-level value 261 for each pixel. In particular, the image data input provided to each lookup table 245, 247,
249 and 251 includes the original grey- level for each pixel within the image frame. Stated another way, each lookup table 245, 247, 249 and 251 receives the original grey- level image data input for each pixel within the image frame.
The outputs 263, 265, 267 and 269 of the lookup tables 245, 247, 249 and 251 provide new grey-level values for each pixel within the corresponding zone. This new grey-level values can be increased, decrease, or the same as the original luminance.
The outputs 263, 265, 267 and 269, of the lookup tables are multiplied by a corresponding position weighting factor 271, 273, 275 and 277. In particular, the new grey-level values for each pixel is multiplied by a weighting factor 271, 273, 275 and 277, based upon the pixel's location relative to one or more reference positions within the image frame. As will be appreciated by those having ordinary skill in the art, the weighting factors 271, 273, 275 and 277, govern how the grey-level within each zone effects the new grey-level values provided for each pixel within the image. Preferably, a pixel is effected more by pixel values within the zone within which it is located, and is effected less dramatically by pixel values from a zone furthest from the pixel. The weightings can be normalized so their total sum is 1.
In one embodiment, the reference positions within the image frame are located at the corners of the image frame. In a further embodiment, the reference positions can be at the centers of the defined zones.
The weightings are based upon the distance the pixel is located from the reference positions. For example, if the image frame is a square containing four equally sized zones with the reference positions at the corners of the image frame, then the weighting factors 271, 273, 275 and 277, for the center pixel within the image frame would be 0.25 because the pixel is equal distant from each zone's reference position (i.e., the corners of the image frame). Likewise, each pixel at the corners of the image frame would be weighted such that only the lookup table results for the zone containing the pixel would be applied in determining the pixel's new grey-level. In yet another embodiment, and as will be appreciated by those having ordinary skill in the art, only a single zone can be used within an image frame. Preferably, the zone is centered in the image frame and is small in size than the entire image frame.
It should be understood that the device 10 and its combination of elements, and derivatives thereof, provide circuits and implement techniques that are highly efficient to implementation yet work very well. Others have proposed various forms of contrast enhancement solutions but the proposals are much more complex to implement. Apparatus and methods according to the invention create an algorithmically efficient procedure that results in cost efficient real-time implementation. This, in turn provides lower product costs.
While the specific embodiments have been illustrated and described, numerous modifications come to mind without significantly departing from the spirit of the invention and the scope of protection is only limited by the scope of the accompanying Claims.

Claims

1. An apparatus providing an image output signal in response to an image input signal, the device comprising: a saturation bias identification circuit having a range of useful grey-levels output responsive to the image input signal; and a cumulative distribution function scaling circuit having a scaled output responsive to the useful grey-levels output.
2. The apparatus of Claim 1 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
3. The apparatus of Claim 1 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
4. The apparatus of Claim 1 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
5. The apparatus of Claim 1 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (ko) plus X additional grey-levels through the last unsaturated grey-level (k„) minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
6. The apparatus of Claim 1 further comprising a video decoder having a digital output responsive to the image input signal, and wherein the range of useful grey- levels output is responsive to the digital output of the video decoder.
7. The apparatus of Claim 1 further comprising a video encoder having an analog output responsive to the image input signal.
8. The apparatus of Claim 1 further comprising a logical device containing the saturation bias identification circuit.
9. The apparatus of Claim 8 wherein the logical device is a field programmable gate array.
10. The apparatus of Claim 1 further comprising a color transform having a luminance output signal responsive to the image input signal.
11. The apparatus of Claim 1 further comprising an image pre-contrast conditioner having a conditioned output responsive to the image input signal.
12. The apparatus of Claim 11 wherein the image pre-contrast conditioner includes a median filter.
13. The apparatus of Claim 11 wherein the image pre-contrast conditioner includes a Gaussian filter.
14. The apparatus of Claim 11 wherein the image pre-contrast conditioner includes a Laplace filter.
15. The apparatus of Claim 12 wherein the image pre-contrast conditioner includes another median filter.
16. The apparatus of Claim 1 further comprising a sample area selection circuit having a selected image output signal responsive to the image input signal and an area selection input signal.
17. The apparatus of Claim 1 further comprising a histogram accumulation circuit having a histogram output responsive to the image input signal.
18. The apparatus of Claim 1 further comprising a cumulative distribution circuit having a cumulative distribution output responsive to the image input signal.
19. The apparatus of Claim 1 further comprising a lookup construction circuit having a lookup table output responsive to the linear scaled output.
20. The apparatus of Claim 19 further comprising an enhanced luminance image generation circuit having an enhanced image output responsive to the lookup table and the image input signal.
21. The apparatus of Claim 1 further comprising a switch for selecting the range of useful grey-levels.
22. A method of providing an image output signal in response to an image input signal comprising the steps of: generating a range of useful grey-levels in response to the image input signal; and generating a scaled output in response to the range of useful grey-levels.
23. The method of Claim 22 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
24. The method of Claim 22 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
25. The method of Claim 22 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
26. The method of Claim 22 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (kg) plus X additional grey-levels through the last unsaturated grey-level (kn) minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
27. The method of Claim 22 further comprising the step of generating a digital output responsive to the image input signal.
28. The method of Claim 22 further comprising the step of generating an analog output in response to the image input signal.
29. The method of Claim 22 further comprising the step of generating a luminance output signal in response to the image input signal.
30. The method of Claim 22 further comprising the step of generating a conditioned output responsive to the image input signal.
31. The method of Claim 30 wherein the step of generating a conditioned output includes the step of applying a median filter to a pixel array.
32. The method of Claim 30 wherein the step of generating a conditioned output includes the step of applying a Gaussian filter to a pixel array.
33. The method of Claim 30 wherein the step of generating a conditioned output includes the step of applying a Laplace filter to a pixel array.
34. The method of Claim 31 further comprising the step of applying another median filter to a filtered pixel array.
35. The method of Claim 31 further comprising the step of generating a selected image output signal in response to the image input signal and an area selection input signal.
36. The method of Claim 31 further comprising the step of generating a histogram output in response to the image input signal.
37. The method of Claim 31 further comprising the step of generating a cumulative distribution output in response to the image input signal.
38. The method of Claim 31 further comprising the step of generating a lookup table in response to the linear scaled output.
39. The method of Claim 31 further comprising the step of generating an enhanced image output in response to the lookup table and the image input signal.
40. The method of Claim 22 further comprising the step of selecting the range of useful grey-levels.
41. A device providing an image output frame in response to an image input frame, the device comprising: a housing having a printed circuit board contained therein; an input connector attached to the housing and having a conductive path attached to the printed circuit board; an output connector attached to the housing and having a conductive path attached to the printed circuit board; an integrated circuit placed on the printed circuit board, the integrated circuit having an output responsive to the image input frame, the output comprising transformed pixel frame data; and wherein the device does not require a keyboard to operate.
42. The device of Claim 41 wherein the device is an embedded system.
43. The device of Claim 41 wherein the input connector receives thirty frames per second and the output connector provides thirty frames per second.
44. The device of Claim 41 wherein a port is not provided for operably attaching the device to the keyboard.
45. The device of Claim 41 , further comprising a saturation bias identification circuit having a range of useful grey-levels output responsive to the image input frame, and a cumulative distribution function scaling circuit having a scaled output responsive to the useful grey-levels output.
46. The device of Claim 45 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
47. The device of Claim 45 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
48. The device of Claim 45 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
49. The device of Claim 45 wherein the range of useful grey-levels output includes all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (ko) plus X additional grey-levels through the last unsaturated grey-level (kj minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
50. The device of Claim 45 further comprising a video decoder having a digital output responsive to the image input frame, and wherein the range of useful grey- levels output is responsive to the digital output of the video decoder.
51. The device of Claim 45 further comprising a video encoder having an analog output responsive to the image input frame.
52. The device of Claim 45 further comprising a logical device containing the saturation bias identification circuit.
53. The device of Claim 52 wherein the logical device is a field programmable gate array.
54. The device of Claim 45 further comprising a color transform having a luminance output frame responsive to the image input frame.
55. The device of Claim 45 further comprising an image pre-contrast conditioner having a conditioned output responsive to the image input frame.
56. The device of Claim 55 wherein the image pre-contrast conditioner includes a median filter.
57. The device of Claim 55 wherein the image pre-contrast conditioner includes a Gaussian filter.
58. The device of Claim 55 wherein the image pre-contrast conditioner includes a
Laplace filter.
59. The device of Claim 56 wherein the image pre-contrast conditioner includes another median filter.
60. The device of Claim 45 further comprising a sample area selection circuit having a selected image output frame responsive to the image input frame and an area selection input frame.
61. The device of Claim 45 further comprising a histogram accumulation circuit having a histogram output responsive to the image input frame.
62. The device of Claim 45 further comprising a cumulative distribution circuit having a cumulative distribution output responsive to the image input frame.
63. The device of Claim 45 further comprising a lookup construction circuit having a lookup table output responsive to the linear scaled output.
64. The device of Claim 63 further comprising an enhanced luminance image generation circuit having an enhanced image output responsive to the lookup table and the image input frame.
65. The device of Claim 45 wherein the range of useful-grey levels is responsive to a switch position.
66. A method for providing an image output frame in response to an image input frame, the method comprising the steps of: segmenting the image input frame into one or more zones; determining a plurality of grey-level values for a pixel based, at least in part, on grey-level data contained within the one or more zones; and, calculating a composite enhanced pixel grey-level value for the pixel by blending the plurality of grey-level values.
67. The method of Claim 66 wherein the image input frame is segmented into a plurality of zones.
68. The method of Claim 66 further comprising establishing reference points within the image input frame, and wherein the step of calculating a composite enhanced pixel grey-level value is based, at least in part, on distance from the reference points.
69. The method of Claim 67 further comprising generating a range of useful grey- levels for each zone, and generating a scaled output in response to at least one of the ranges of useful grey-levels.
70. The method of Claim 69 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
71. The method of Claim 69 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
72. The method of Claim 69 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
73. The method of Claim 69 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (ko) plus X additional grey-levels through the last unsaturated grey-level (k_) minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
74. The method of Claim 66 further comprising the step of generating a digital output responsive to the image input frame.
75. The method of Claim 66 further comprising the step of generating an analog output in response to the image input frame.
76. The method of Claim 66 further comprising the step of generating a luminance output signal in response to the image input frame.
77. The method of Claim 66 further comprising the step of generating a conditioned output responsive to the image input frame.
78. The method of Claim 77 wherein the step of generating a conditioned output includes the step of applying a median filter to a pixel array.
79. The method of Claim 77 wherein the step of generating a conditioned output includes the step of applying a Gaussian filter to a pixel array.
80. The method of Claim 77 wherein the step of generating a conditioned output includes the step of applying a Laplace filter to a pixel array.
81. The method of Claim 78 further comprising the step of applying another median filter to a filtered pixel array.
82. The method of Claim 66 further comprising the step of generating a selected image output signal in response to the image input frame and an area selection input signal.
83. The method of Claim 66 further comprising the step of generating a histogram output in response to the image input frame.
84. The method of Claim 66 further comprising the step of generating a cumulative distribution output in response to the image input frame.
85. The method of Claim 69 further comprising the step of generating a lookup table in response to the scaled output.
86. The method of Claim 85 further comprising the step of generating an enhanced image output in response to the lookup table and the image input signal.
87. The method of Claim 66 further comprising the step of segmenting the image input frame into four zones with each zone comprising a mutually exclusive quadrant of the image input frame.
88. A method for providing an image output frame in response to an image input frame, the method comprising the steps of: establishing reference points within the image input frame; calculating grey-level values for a plurality of pixels within the image input frame; and, calculating a composite enhanced pixel grey-level value for the pixels based, at least in part, on distance from the reference points.
89. The method of Claim 88 further comprising segmenting the image input frame into a plurality of zones, and wherein at least one of the reference points is located within one of the zones.
90. The method of Claim 88 further comprising generating a range of useful grey- levels in response to the image input frame, and generating a scaled output in response to the range of useful grey-levels.
91. The method of Claim 90 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
92. The method of Claim 90 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
93. The method of Claim 90 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
94. The method of Claim 90 wherein the step of generating a range of useful grey- levels includes specifying all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (kg) plus X additional grey-levels through the last unsaturated grey-level (k„) minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
95. The method of Claim 88 further comprising the step of generating a digital output responsive to the image input frame.
96. The method of Claim 88 further comprising the step of generating an analog output in response to the image input frame.
97. The method of Claim 88 further comprising the step of generating a luminance output signal in response to the image input frame.
98. The method of Claim 88 further comprising the step of generating a conditioned output responsive to the image input frame.
99. The method of Claim 98 wherein the step of generating a conditioned output includes the step of applying a median filter to a pixel array.
100. The method of Claim 98 wherein the step of generating a conditioned output includes the step of applying a Gaussian filter to a pixel array.
101. The method of Claim 98 wherein the step of generating a conditioned output includes the step of applying a Laplace filter to a pixel array.
102. The method of Claim 99 further comprising the step of applying another median filter to a filtered pixel array.
103. The method of Claim 99 further comprising the step of generating a selected image output signal in response to the image input frame and an area selection input signal.
104. The method of Claim 99 further comprising the step of generating a histogram output in response to the image input frame.
105. The method of Claim 99 further comprising the step of generating a cumulative distribution output in response to the image input frame.
106. The method of Claim 88 further comprising the step of segmenting the image input frame into four zones with each zone comprising a mutually exclusive quadrant of the image input frame.
107. A method for providing an image output frame in response to an image input frame, the method comprising the steps of: constructing an equalized lookup table for the image input frame; constructing an equalized lookup table for a zone within the image input frame; and, utilizing the lookup tables to build a balanced lookup table.
108. The method of Claim 107 further comprising the step of utilizing a lookup table representative of the image input frame to build the balanced lookup table.
109. The method of Claim 108 wherein the lookup table representative of the image input frame is an identity lookup table.
110. The method of Claim 107 further comprising the step of segmenting the image input frame into a plurality of zones.
111. The method of Claim 110 further comprising establishing reference points within at least two of the zones, and calculating a composite enhanced pixel grey- level value based, at least in part, on distance from the reference points.
112. The method of Claim 107 wherein the step of constructing an equalized lookup table for a zone includes generating a range of useful grey-levels within the zone, and generating a scaled output in response to the range of useful grey-levels.
113. The method of Claim 112 wherein the step of generating a range of useful grey-levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function.
114. The method of Claim 112 wherein the step of generating a range of useful grey-levels includes specifying all grey-levels within a cumulative distribution function except the last unsaturated grey-level within the cumulative distribution function.
115. The method of Claim 112 wherein the step of generating a range of useful grey-levels includes specifying all grey-levels within a cumulative distribution function except the first unsaturated grey-level within the cumulative distribution function and the last unsaturated grey-level within the cumulative distribution function.
116. The method of Claim 112 wherein the step of generating a range of useful grey-levels includes specifying all grey-levels within a cumulative distribution function in the range from, and including, the first unsaturated grey-level (ko) plus X additional grey-levels through the last unsaturated grey-level (k minus Y additional grey-levels, wherein X and Y are whole numbers greater than zero.
117. The method of Claim 112 further comprising the step of generating a digital output responsive to the image input frame.
118. The method of Claim 107 further comprising the step of generating an analog output in response to the image input frame.
119. The method of Claim 107 further comprising the step of generating a luminance output signal in response to the image input frame.
120. The method of Claim 107 further comprising the step of generating a conditioned output responsive to the image input frame.
121. The method of Claim 120 wherein the step of generating a conditioned output includes the step of applying a median filter to a pixel array.
122. The method of Claim 120 wherein the step of generating a conditioned output includes the step of applying a Gaussian filter to a pixel array.
123. The method of Claim 120 wherein the step of generating a conditioned output includes the step of applying a Laplace filter to a pixel array.
124. The method of Claim 107 further comprising the step of segmenting the image input frame into four zones with each of said zones comprising a mutually exclusive quadrant of the image input frame.
125. An apparatus providing an output of signal values in response to an input of signal values derived directly or indirectly from a sensor, wherein each input value is variable within a range of values based upon an environment sensed by the sensor, and each value of the input signal has a spatial and or geometric relationship to other values in the input signal, the device comprising: a signal intensity over limit or saturation identification circuit having a range of useful signal intensities output responsive to the image input signal; and a cumulative distribution function scaling circuit having a scaled output responsive to useful signal intensity output.
126. A method of providing an output of signal values in response to an input of signal values derived directly or indirectly from a sensor, wherein each input value is variable within a range of values based upon an environment sensed by the sensor, and each value of the input signal has a spatial and or geometric relationship to other values in the input signal, comprising the steps of: generating a range of useful signal intensities in response to the image input signal; and generating a scaled output in response to the range of useful signal intensities.
127. A method for providing an output of signal values in response to an input array of signal values derived directly or indirectly from a sensor, wherein each input value is variable within a range of values based upon an environment sensed by the sensor, and each value of the input signal has coordinates with a spatial and or geometric relationship to other values in the input signal, comprising the steps of: segmenting the image input array into one or more spatial zones; determining a plurality of signal intensity values for a coordinate based, at least in part, on signal intensity contained at coordinates within the one or more other zones; and, calculating a composite enhanced coordinate signal value for the coordinate by blending the plurality of signal intensity values.
128. A method for providing an output of signal values in response to an input array of signal values derived directly or indirectly from a sensor, wherein each input value is variable within a range of values based upon an environment sensed by the sensor, and each value of the input signal has coordinates with a spatial and or geometric relationship to other values in the input signal, comprising the steps of: establishing reference points within the array of signal values; calculating signal intensity values for a plurality of coordinates within the array of input values; and, calculating a composite enhanced coordinate signal intensity value for the coordinates based, at least in part, on a distance of the coordinate from the reference points.
129. A method providing an output of signal values in response to an input array of signal values derived directly or indirectly from a sensor, wherein each input value is variable within a range of values based upon an environment sensed by the sensor, and each value of the input signal has coordinates with a spatial and or geometric relationship to other values in the input signal, comprising the steps of: constructing an equalized lookup table for the array of signal intensity values; constructing an equalized lookup table for a zone within the array of signal intensity values; and, utilizing the lookup tables to build a balanced lookup table.
PCT/US2003/028028 2002-09-06 2003-09-08 Signal intensity range transformation apparatus and method WO2004023787A2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2003270386A AU2003270386A1 (en) 2002-09-06 2003-09-08 Signal intensity range transformation apparatus and method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US40866302P 2002-09-06 2002-09-06
US60/408,663 2002-09-06

Publications (2)

Publication Number Publication Date
WO2004023787A2 true WO2004023787A2 (en) 2004-03-18
WO2004023787A3 WO2004023787A3 (en) 2004-05-13

Family

ID=31978654

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2003/028028 WO2004023787A2 (en) 2002-09-06 2003-09-08 Signal intensity range transformation apparatus and method

Country Status (3)

Country Link
US (2) US7321699B2 (en)
AU (1) AU2003270386A1 (en)
WO (1) WO2004023787A2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1881695A1 (en) * 2005-05-13 2008-01-23 Olympus Corporation Image processing apparatus, and image processing program
WO2013038333A3 (en) * 2011-09-18 2013-05-23 Forus Health Pvt. Ltd. Method and system for enhancing image quality
CN107533755A (en) * 2015-04-14 2018-01-02 皇家飞利浦有限公司 For improving the apparatus and method of medical image quality

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6937274B2 (en) * 2001-12-17 2005-08-30 Motorola, Inc. Dynamic range compression of output channel data of an image sensor
KR100513273B1 (en) * 2003-07-04 2005-09-09 이디텍 주식회사 Apparatus and method for real-time brightness control of moving images
JP4428159B2 (en) * 2003-11-05 2010-03-10 セイコーエプソン株式会社 Image data generation apparatus, image quality correction apparatus, image data generation method, and image quality correction method
US7355639B2 (en) * 2003-11-06 2008-04-08 Omnivision Technologies, Inc. Lens correction using processed YUV data
US20060008174A1 (en) * 2004-07-07 2006-01-12 Ge Medical Systems Global Technology Count adaptive noise reduction method of x-ray images
JP4533330B2 (en) * 2005-04-12 2010-09-01 キヤノン株式会社 Image forming apparatus and image forming method
US7859574B1 (en) * 2005-07-19 2010-12-28 Maxim Integrated Products, Inc. Integrated camera image signal processor and video encoder
WO2007016397A2 (en) * 2005-08-01 2007-02-08 Bioptigen, Inc. Methods, systems and computer programm for 3d-registrati0n of three dimensional data sets obtained by preferably optical coherence tomography based on the alignment of projection images or fundus images, respectively
JP2007060169A (en) * 2005-08-23 2007-03-08 Sony Corp Image processing apparatus, image display apparatus and image processing method
US7512574B2 (en) * 2005-09-30 2009-03-31 International Business Machines Corporation Consistent histogram maintenance using query feedback
US7562169B2 (en) * 2005-12-30 2009-07-14 Symwave, Inc. Video and audio front end assembly and method
US7894686B2 (en) * 2006-01-05 2011-02-22 Lsi Corporation Adaptive video enhancement gain control
KR100717401B1 (en) * 2006-03-02 2007-05-11 삼성전자주식회사 Method and apparatus for normalizing voice feature vector by backward cumulative histogram
KR100849845B1 (en) * 2006-09-05 2008-08-01 삼성전자주식회사 Method and apparatus for Image enhancement
US7912307B2 (en) * 2007-02-08 2011-03-22 Yu Wang Deconvolution method using neighboring-pixel-optical-transfer-function in fourier domain
CN101784907A (en) * 2007-04-30 2010-07-21 皇家飞利浦电子股份有限公司 Positive contrast MR susceptibility imaging
US8744159B2 (en) * 2010-03-05 2014-06-03 Bioptigen, Inc. Methods, systems and computer program products for collapsing volume data to lower dimensional representations thereof using histogram projection
US9866899B2 (en) 2012-09-19 2018-01-09 Google Llc Two way control of a set top box
US9788055B2 (en) 2012-09-19 2017-10-10 Google Inc. Identification and presentation of internet-accessible content associated with currently playing television programs
US8907973B2 (en) * 2012-10-22 2014-12-09 Stmicroelectronics International N.V. Content adaptive image restoration, scaling and enhancement for high definition display
US9646366B2 (en) * 2012-11-30 2017-05-09 Change Healthcare Llc Method and apparatus for enhancing medical images
US9377291B2 (en) 2013-12-05 2016-06-28 Bioptigen, Inc. Image registration, averaging, and compounding for high speed extended depth optical coherence tomography
KR102221116B1 (en) * 2015-02-27 2021-02-26 에스케이하이닉스 주식회사 A device and method for removing the noise on the image using cross-kernel type median filter
CN104900209A (en) * 2015-06-29 2015-09-09 深圳市华星光电技术有限公司 Overdriven target value calculating method based on sub-pixel signal bright-dark switching
CN111050167B (en) * 2016-11-04 2021-11-09 谷歌有限责任公司 Method and apparatus for recovering degraded frames resulting from reconstruction of source frames
EP3571843A1 (en) * 2017-01-18 2019-11-27 Dolby Laboratories Licensing Corporation Segment-based reshaping for coding high dynamic range video
WO2022015985A1 (en) 2020-07-15 2022-01-20 Copado, Inc. Methods for software development and operation process analytics and devices thereof
US11775910B2 (en) 2020-07-15 2023-10-03 Copado, Inc. Applied computer technology for high efficiency value stream management and mapping and process tracking
CN114972218B (en) * 2022-05-12 2023-03-24 中海油信息科技有限公司 Pointer meter reading identification method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5793883A (en) * 1995-09-29 1998-08-11 Siemens Medical Systems, Inc. Method for enhancing ultrasound image
US5949918A (en) * 1997-05-21 1999-09-07 Sarnoff Corporation Method and apparatus for performing image enhancement
US6078686A (en) * 1996-09-30 2000-06-20 Samsung Electronics Co., Ltd. Image quality enhancement circuit and method therefor
US6507668B1 (en) * 1998-12-15 2003-01-14 Samsung Electronics Co., Ltd. Image enhancing apparatus and method of maintaining brightness of input image

Family Cites Families (172)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US2758676A (en) 1954-02-16 1956-08-14 Haughton Elevator Company Variable standing time control
US3195126A (en) 1957-05-13 1965-07-13 Lab For Electronics Inc Traffic supervisory system
US3686434A (en) 1957-06-27 1972-08-22 Jerome H Lemelson Area surveillance system
US3255434A (en) 1961-11-01 1966-06-07 Peter D Schwarz Vehicle detection and counting system
US3590151A (en) 1966-12-30 1971-06-29 Jackson & Church Electronics C Television surveillance system
US3562423A (en) 1967-08-15 1971-02-09 Univ Northwestern Dictorial tracking and recognition system which provides display of target identified by brilliance and spatial characteristics
US3924130A (en) 1968-02-12 1975-12-02 Us Navy Body exposure indicator
US3534499A (en) 1969-01-27 1970-10-20 John L Czinger Jr Door opening apparatus
US3685012A (en) 1970-04-16 1972-08-15 Sperry Rand Corp Apparatus for determining data associated with objects
US3691556A (en) 1970-06-03 1972-09-12 Memco Electronics Ltd Detection of movement in confined spaces
US3663937A (en) 1970-06-08 1972-05-16 Thiokol Chemical Corp Intersection ingress-egress automatic electronic traffic monitoring equipment
US3668625A (en) 1970-09-21 1972-06-06 David Wolf Monitoring system
US3740466A (en) 1970-12-14 1973-06-19 Jackson & Church Electronics C Surveillance system
US3691302A (en) 1971-02-25 1972-09-12 Gte Sylvania Inc Automatic light control for low light level television camera
GB1378754A (en) 1971-09-07 1974-12-27 Peak Technologies Ltd Patient monitoring
US3816648A (en) 1972-03-13 1974-06-11 Magnavox Co Scene intrusion alarm
US3890463A (en) 1972-03-15 1975-06-17 Konan Camera Res Inst System for use in the supervision of a motor-boat race or a similar timed event
US3852592A (en) 1973-06-07 1974-12-03 Stanley Works Automatic door operator
US3988533A (en) 1974-09-30 1976-10-26 Video Tek, Inc. Video-type universal motion and intrusion detection system
US3947833A (en) 1974-11-20 1976-03-30 The United States Of America As Represented By The Secretary Of The Navy Automatic target detection system
US3930735A (en) 1974-12-11 1976-01-06 The United States Of America As Represented By The United States National Aeronautics And Space Administration Traffic survey system
JPS5197155A (en) 1975-02-21 1976-08-26 Erebeetano jokyakudeetashushusochi
FR2321229A1 (en) 1975-08-13 1977-03-11 Cit Alcatel METHOD AND APPARATUS FOR AUTOMATIC GRAPHIC CONTROL
SE394146B (en) 1975-10-16 1977-06-06 L Olesen SATURATION DEVICE RESP CONTROL OF A FOREMAL, IN ESPECIALLY THE SPEED OF A VEHICLE.
DE2617111C3 (en) 1976-04-17 1986-02-20 Robert Bosch Gmbh, 7000 Stuttgart Method for detecting movement in the surveillance area of a television camera
DE2617112C3 (en) 1976-04-17 1982-01-14 Robert Bosch Gmbh, 7000 Stuttgart Method for determining a movement or a change in the surveillance area of a television camera
US4063282A (en) 1976-07-20 1977-12-13 The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration TV fatigue crack monitoring system
DE2638138C3 (en) 1976-08-25 1979-05-03 Kloeckner-Werke Ag, 4100 Duisburg Device for recognizing and sorting out defective packs that are transported along a conveyor line
US4240109A (en) 1976-10-14 1980-12-16 Micro Consultants, Limited Video movement detection
US4136950A (en) 1976-11-08 1979-01-30 Labrum Engineering, Inc. Microscope system for observing moving particles
US4183013A (en) 1976-11-29 1980-01-08 Coulter Electronics, Inc. System for extracting shape features from an image
JPS53110823A (en) 1977-03-10 1978-09-27 Ricoh Co Ltd Optical information processor
DE2715083C3 (en) 1977-04-04 1983-02-24 Robert Bosch Gmbh, 7000 Stuttgart System for the discrimination of a video signal
US4141062A (en) * 1977-05-06 1979-02-20 Trueblood, Inc. Trouble light unit
DE2720865A1 (en) 1977-05-10 1978-11-23 Philips Patentverwaltung ARRANGEMENT FOR THE EXAMINATION OF OBJECTS
US4163357A (en) 1977-06-13 1979-08-07 Hamel Gmbh, Zwirnmaschinen Apparatus for cable-twisting two yarns
US4133004A (en) 1977-11-02 1979-01-02 Hughes Aircraft Company Video correlation tracker
JPS5474700A (en) 1977-11-26 1979-06-14 Agency Of Ind Science & Technol Collection and delivery system for traffic information by photo electric conversion element group
CA1103803A (en) 1978-03-01 1981-06-23 National Research Council Of Canada Method and apparatus of determining the center of area or centroid of a geometrical area of unspecified shape lying in a larger x-y scan field
JPS54151322A (en) 1978-05-19 1979-11-28 Tokyo Hoso:Kk Storoboscopic effect generator for television
US4187519A (en) 1978-08-17 1980-02-05 Rockwell International Corporation System for expanding the video contrast of an image
CA1116286A (en) 1979-02-20 1982-01-12 Control Data Canada, Ltd. Perimeter surveillance system
US4257063A (en) 1979-03-23 1981-03-17 Ham Industries, Inc. Video monitoring system and method
US4219845A (en) 1979-04-12 1980-08-26 The United States Of America As Represented By The Secretary Of The Air Force Sense and inject moving target indicator apparatus
DE2934038C2 (en) 1979-08-23 1982-02-25 Deutsche Forschungs- und Versuchsanstalt für Luft- und Raumfahrt e.V., 5000 Köln Crack propagation measuring device
US4414685A (en) 1979-09-10 1983-11-08 Sternberg Stanley R Method and apparatus for pattern recognition and detection
US4395699A (en) 1979-09-10 1983-07-26 Environmental Research Institute Of Michigan Method and apparatus for pattern recognition and detection
US4317130A (en) 1979-10-10 1982-02-23 Motorola, Inc. Narrow band television transmission system
JPS56132505A (en) 1980-03-24 1981-10-16 Hitachi Ltd Position detecting method
US4298858A (en) 1980-03-27 1981-11-03 The United States Of America As Represented By The Secretary Of The Air Force Method and apparatus for augmenting binary patterns
JPS56160183A (en) 1980-05-09 1981-12-09 Hajime Sangyo Kk Method and device for monitoring
US4337481A (en) 1980-06-10 1982-06-29 Peter Mick Motion and intrusion detecting system
US4410910A (en) 1980-09-18 1983-10-18 Advanced Diagnostic Research Corp. Motion detecting method and apparatus
US4433325A (en) 1980-09-30 1984-02-21 Omron Tateisi Electronics, Co. Optical vehicle detection system
JPS6360960B2 (en) 1980-10-22 1988-11-25
US4455550A (en) 1980-11-06 1984-06-19 General Research Of Electronics, Inc. Detection circuit for a video intrusion monitoring apparatus
JPS57125496A (en) 1981-01-27 1982-08-04 Fujitec Kk Condition detector
US4520343A (en) 1980-12-16 1985-05-28 Hiroshi Koh Lift control system
JPH0210077B2 (en) 1981-01-26 1990-03-06 Memuko Metsudo Ltd
US4493420A (en) 1981-01-29 1985-01-15 Lockwood Graders (U.K.) Limited Method and apparatus for detecting bounded regions of images, and method and apparatus for sorting articles and detecting flaws
DE3107901A1 (en) 1981-03-02 1982-09-16 Siemens AG, 1000 Berlin und 8000 München DIGITAL REAL-TIME TELEVISION IMAGE DEVICE
US4449144A (en) 1981-06-26 1984-05-15 Omron Tateisi Electronics Co. Apparatus for detecting moving body
US4479145A (en) 1981-07-29 1984-10-23 Nippon Kogaku K.K. Apparatus for detecting the defect of pattern
US4433438A (en) 1981-11-25 1984-02-21 The United States Of America As Represented By The Secretary Of The Air Force Sobel edge extraction circuit for image processing
US4589139A (en) 1982-02-04 1986-05-13 Nippon Kogaku K. K. Apparatus for detecting defects in pattern
US4490851A (en) 1982-04-16 1984-12-25 The United States Of America As Represented By The Secretary Of The Army Two-dimensional image data reducer and classifier
US4520504A (en) 1982-07-29 1985-05-28 The United States Of America As Represented By The Secretary Of The Air Force Infrared system with computerized image display
US4569078A (en) 1982-09-17 1986-02-04 Environmental Research Institute Of Michigan Image sensor
JPS5994045A (en) 1982-11-22 1984-05-30 Toshiba Corp Image input apparatus
US4577344A (en) 1983-01-17 1986-03-18 Automatix Incorporated Vision system
US4543567A (en) 1983-04-14 1985-09-24 Tokyo Shibaura Denki Kabushiki Kaisha Method for controlling output of alarm information
US4574393A (en) 1983-04-14 1986-03-04 Blackwell George F Gray scale image processor
US4556900A (en) 1983-05-25 1985-12-03 Rca Corporation Scaling device as for quantized B-Y signal
US4665554A (en) 1983-07-13 1987-05-12 Machine Vision International Corporation Apparatus and method for implementing dilation and erosion transformations in digital image processing
KR910009880B1 (en) 1983-07-25 1991-12-03 가부시기가이샤 히다찌세이사꾸쇼 Image motion detecting circuit of interlacing television signal
US4589030A (en) 1983-07-25 1986-05-13 Kley Victor B Solid state camera
IL69327A (en) 1983-07-26 1986-11-30 Elscint Ltd Automatic misregistration correction
CA1246734A (en) 1983-09-08 1988-12-13 Nec Corp Apparatus for detecting movement in a television signal
US4555724A (en) 1983-10-21 1985-11-26 Westinghouse Electric Corp. Elevator system
US4698937A (en) 1983-11-28 1987-10-13 The Stanley Works Traffic responsive control system for automatic swinging door
US4565029A (en) 1983-11-28 1986-01-21 The Stanley Works Traffic responsive control system for automatic swinging door
US4669218A (en) 1984-03-08 1987-06-02 The Stanley Works Traffic responsive control system
US4694329A (en) 1984-04-09 1987-09-15 Corporate Communications Consultants, Inc. Color correction system and method with scene-change detection
US4653109A (en) 1984-07-30 1987-03-24 Lemelson Jerome H Image analysis system and method
US4641356A (en) 1984-08-24 1987-02-03 Machine Vision International Corporation Apparatus and method for implementing dilation and erosion transformations in grayscale image processing
FI70651C (en) 1984-10-05 1986-09-24 Kone Oy OVERHEAD FREQUENCY FOR OIL FITTINGS
US4679077A (en) 1984-11-10 1987-07-07 Matsushita Electric Works, Ltd. Visual Image sensor system
DE3513833A1 (en) 1984-11-14 1986-05-22 Karl-Walter Prof. Dr.-Ing. 5910 Kreuztal Bonfig FUSE PROTECTION INSERT WITH OPTOELECTRICAL DISPLAY DEVICE
US4685145A (en) 1984-12-07 1987-08-04 Fingermatrix, Inc. Conversion of an image represented by a field of pixels in a gray scale to a field of pixels in binary scale
US4680704A (en) 1984-12-28 1987-07-14 Telemeter Corporation Optical sensor apparatus and method for remotely monitoring a utility meter or the like
US4662479A (en) 1985-01-22 1987-05-05 Mitsubishi Denki Kabushiki Kaisha Operating apparatus for elevator
US4739401A (en) 1985-01-25 1988-04-19 Hughes Aircraft Company Target acquisition system and method
GB8518803D0 (en) 1985-07-25 1985-08-29 Rca Corp Locating target patterns within images
US4779131A (en) 1985-07-26 1988-10-18 Sony Corporation Apparatus for detecting television image movement
US4697594A (en) * 1985-08-21 1987-10-06 North American Philips Corporation Displaying a single parameter image
JPS6278979A (en) 1985-10-02 1987-04-11 Toshiba Corp Picture processor
GB2183878B (en) 1985-10-11 1989-09-20 Matsushita Electric Works Ltd Abnormality supervising system
JPH0744689B2 (en) 1985-11-22 1995-05-15 株式会社日立製作所 Motion detection circuit
JPH0766446B2 (en) 1985-11-27 1995-07-19 株式会社日立製作所 Method of extracting moving object image
US5187747A (en) * 1986-01-07 1993-02-16 Capello Richard D Method and apparatus for contextual data enhancement
US4825393A (en) * 1986-04-23 1989-04-25 Hitachi, Ltd. Position measuring method
US4760607A (en) 1986-07-31 1988-07-26 Machine Vision International Corporation Apparatus and method for implementing transformations in grayscale image processing
JPS63222589A (en) * 1987-03-12 1988-09-16 Toshiba Corp Noise reducing circuit
US4823010A (en) * 1987-05-11 1989-04-18 The Stanley Works Sliding door threshold sensor
US4906940A (en) * 1987-08-24 1990-03-06 Science Applications International Corporation Process and apparatus for the automatic detection and extraction of features in images and displays
US4799243A (en) * 1987-09-01 1989-01-17 Otis Elevator Company Directional people counting arrangement
JPH0695008B2 (en) * 1987-12-11 1994-11-24 株式会社東芝 Monitoring device
JPH0672770B2 (en) * 1988-02-01 1994-09-14 豊田工機株式会社 Robot object recognition device
DE68918886T2 (en) * 1988-04-08 1995-06-01 Dainippon Screen Mfg Process for obtaining the outline of an object in an image.
DE3818534A1 (en) * 1988-05-31 1989-12-07 Brunner Wolfgang METHOD FOR PRESENTING THE LOCALLY DISTRIBUTED DISTRIBUTION OF PHYSICAL SIZES ON A DISPLAY, AND DEVICE FOR CARRYING OUT THE METHOD
ATE89362T1 (en) * 1988-06-03 1993-05-15 Inventio Ag METHOD AND DEVICE FOR CONTROLLING THE DOOR POSITION OF AN AUTOMATIC DOOR.
US4985618A (en) * 1988-06-16 1991-01-15 Nicoh Company, Ltd. Parallel image processing system
FR2634551B1 (en) * 1988-07-20 1990-11-02 Siderurgie Fse Inst Rech METHOD AND DEVICE FOR IDENTIFYING THE FINISH OF A METAL SURFACE
US4991092A (en) * 1988-08-12 1991-02-05 The Regents Of The University Of California Image processor for enhancing contrast between subregions of a region of interest
EP0358459B1 (en) * 1988-09-08 1996-06-19 Sony Corporation Picture processing apparatus
JPH0683373B2 (en) * 1988-12-09 1994-10-19 大日本スクリーン製造株式会社 How to set the reference concentration point
US5008739A (en) * 1989-02-13 1991-04-16 Eastman Kodak Company Real-time digital processor for producing full resolution color signals from a multi-color image sensor
JPH0335399A (en) * 1989-06-30 1991-02-15 Toshiba Corp Change area integrating device
JP2953712B2 (en) * 1989-09-27 1999-09-27 株式会社東芝 Moving object detection device
EP0470268B1 (en) * 1990-02-26 1998-06-10 Matsushita Electric Industrial Co., Ltd. Traffic flow change system
KR100204101B1 (en) * 1990-03-02 1999-06-15 가나이 쓰도무 Image processing apparatus
JP2712844B2 (en) * 1990-04-27 1998-02-16 株式会社日立製作所 Traffic flow measurement device and traffic flow measurement control device
EP0548079A1 (en) * 1990-05-25 1993-06-30 Axiom Innovation Limited An image acquisition system
US5305395A (en) * 1990-06-08 1994-04-19 Xerox Corporation Exhaustive hierarchical near neighbor operations on an image
US5319547A (en) * 1990-08-10 1994-06-07 Vivid Technologies, Inc. Device and method for inspection of baggage and other objects
US5596418A (en) * 1990-08-17 1997-01-21 Samsung Electronics Co., Ltd. Deemphasis and subsequent reemphasis of high-energy reversed-spectrum components of a folded video signal
US5182778A (en) * 1990-08-31 1993-01-26 Eastman Kodak Company Dot-matrix video enhancement for optical character recognition
US5181254A (en) * 1990-12-14 1993-01-19 Westinghouse Electric Corp. Method for automatically identifying targets in sonar images
FR2670978A1 (en) * 1990-12-21 1992-06-26 Philips Electronique Lab MOTION EXTRACTION METHOD COMPRISING THE FORMATION OF DIFFERENCE IMAGES AND THREE DIMENSIONAL FILTERING.
JPH04241077A (en) * 1991-01-24 1992-08-28 Mitsubishi Electric Corp Moving body recognizing method
US5296852A (en) * 1991-02-27 1994-03-22 Rathi Rajendra P Method and apparatus for monitoring traffic flow
DE69232732T2 (en) * 1991-03-19 2009-09-10 Mitsubishi Denki K.K. Measuring device for moving bodies and image processing device for measuring traffic flow
JP2936791B2 (en) * 1991-05-28 1999-08-23 松下電器産業株式会社 Gradation correction device
US5509082A (en) * 1991-05-30 1996-04-16 Matsushita Electric Industrial Co., Ltd. Vehicle movement measuring apparatus
JPH0578048A (en) * 1991-09-19 1993-03-30 Hitachi Ltd Detecting device for waiting passenger in elevator hall
US5289520A (en) * 1991-11-27 1994-02-22 Lorad Corporation Stereotactic mammography imaging system with prone position examination table and CCD camera
KR940001054B1 (en) * 1992-02-27 1994-02-08 삼성전자 주식회사 Automatic contrast compensating method and it's apparatus in color video printer
US5500904A (en) * 1992-04-22 1996-03-19 Texas Instruments Incorporated System and method for indicating a change between images
JP2917661B2 (en) * 1992-04-28 1999-07-12 住友電気工業株式会社 Traffic flow measurement processing method and device
US5300739A (en) * 1992-05-26 1994-04-05 Otis Elevator Company Cyclically varying an elevator car's assigned group in a system where each group has a separate lobby corridor
US5612928A (en) * 1992-05-28 1997-03-18 Northrop Grumman Corporation Method and apparatus for classifying objects in sonar images
US5410418A (en) * 1992-06-24 1995-04-25 Dainippon Screen Mfg. Co., Ltd. Apparatus for converting image signal representing image having gradation
US5483351A (en) * 1992-09-25 1996-01-09 Xerox Corporation Dilation of images without resolution conversion to compensate for printer characteristics
US5617484A (en) * 1992-09-25 1997-04-01 Olympus Optical Co., Ltd. Image binarizing apparatus
US5387768A (en) * 1993-09-27 1995-02-07 Otis Elevator Company Elevator passenger detector and door control system which masks portions of a hall image to determine motion and court passengers
US5592567A (en) * 1992-11-10 1997-01-07 Siemens Aktiengesellschaft Method for detecting and separating the shadow of moving objects in a sequence of digital images
EP0607706B1 (en) * 1993-01-01 2000-09-20 Canon Kabushiki Kaisha Image processing apparatus and method
US5282337A (en) * 1993-02-22 1994-02-01 Stanley Home Automation Garage door operator with pedestrian light control
ATE225964T1 (en) * 1993-03-31 2002-10-15 Luma Corp INFORMATION MANAGEMENT IN AN ENDOSCOPY SYSTEM
BE1007608A3 (en) * 1993-10-08 1995-08-22 Philips Electronics Nv Improving picture signal circuit.
US5604822A (en) * 1993-11-12 1997-02-18 Martin Marietta Corporation Methods and apparatus for centroid based object segmentation in object recognition-type image processing system
US5875264A (en) * 1993-12-03 1999-02-23 Kaman Sciences Corporation Pixel hashing image recognition system
JP3721206B2 (en) * 1993-12-24 2005-11-30 富士写真フイルム株式会社 Image reproduction device
US5621868A (en) * 1994-04-15 1997-04-15 Sony Corporation Generating imitation custom artwork by simulating brush strokes and enhancing edges
US5625709A (en) * 1994-12-23 1997-04-29 International Remote Imaging Systems, Inc. Method and apparatus for identifying characteristics of an object in a field of view
US5982926A (en) * 1995-01-17 1999-11-09 At & T Ipm Corp. Real-time image enhancement techniques
US6324558B1 (en) * 1995-02-14 2001-11-27 Scott A. Wilber Random number generator and generation method
US5727080A (en) * 1995-05-03 1998-03-10 Nec Research Institute, Inc. Dynamic histogram warping of image histograms for constant image brightness, histogram matching and histogram specification
FR2734911B1 (en) * 1995-06-01 1997-08-01 Aerospatiale METHOD AND DEVICE FOR DETECTING THE MOVEMENT OF A TARGET AND THEIR APPLICATIONS
US5857029A (en) * 1995-06-05 1999-01-05 United Parcel Service Of America, Inc. Method and apparatus for non-contact signature imaging
US5978506A (en) * 1995-12-28 1999-11-02 Ricoh & Company, Ltd. Colorant-independent color balancing methods and systems
DE69716803T2 (en) * 1996-04-10 2003-03-27 Samsung Electronics Co Ltd Image quality improvement method by histogram equalization with mean agreement and circuit therefor
US6020931A (en) * 1996-04-25 2000-02-01 George S. Sheng Video composition and position system and media signal communication system
US5872857A (en) * 1996-05-22 1999-02-16 Raytheon Company Generalized biased centroid edge locator
JP3602659B2 (en) * 1996-08-05 2004-12-15 株式会社リコー Feature extraction method and feature extraction device from gray value document image
US5859698A (en) * 1997-05-07 1999-01-12 Nikon Corporation Method and apparatus for macro defect detection using scattered light
JP3881439B2 (en) * 1998-01-23 2007-02-14 シャープ株式会社 Image processing device
US6760086B2 (en) * 1998-03-26 2004-07-06 Tomoegawa Paper Co., Ltd. Attachment film for electronic display device
DE69830583T2 (en) * 1998-09-22 2006-06-01 Hewlett-Packard Development Co., L.P., Houston Method and device for processing image data
US6538405B1 (en) * 2000-04-28 2003-03-25 The Cherry Corporation Accessory control system
TW508560B (en) * 2001-04-03 2002-11-01 Chunghwa Picture Tubes Ltd Method for performing different anti-compensation processes by segments on image gray levels inputted to plasma flat display
US7006688B2 (en) * 2001-07-05 2006-02-28 Corel Corporation Histogram adjustment features for use in imaging technologies

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5793883A (en) * 1995-09-29 1998-08-11 Siemens Medical Systems, Inc. Method for enhancing ultrasound image
US6078686A (en) * 1996-09-30 2000-06-20 Samsung Electronics Co., Ltd. Image quality enhancement circuit and method therefor
US5949918A (en) * 1997-05-21 1999-09-07 Sarnoff Corporation Method and apparatus for performing image enhancement
US6507668B1 (en) * 1998-12-15 2003-01-14 Samsung Electronics Co., Ltd. Image enhancing apparatus and method of maintaining brightness of input image

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1881695A1 (en) * 2005-05-13 2008-01-23 Olympus Corporation Image processing apparatus, and image processing program
EP1881695A4 (en) * 2005-05-13 2009-09-16 Olympus Corp Image processing apparatus, and image processing program
US8184924B2 (en) 2005-05-13 2012-05-22 Olympus Corporation Image processing apparatus and image processing program
WO2013038333A3 (en) * 2011-09-18 2013-05-23 Forus Health Pvt. Ltd. Method and system for enhancing image quality
US9002131B2 (en) 2011-09-18 2015-04-07 Forus Health Pvt. Ltd. Method and system for enhancing image quality
CN107533755A (en) * 2015-04-14 2018-01-02 皇家飞利浦有限公司 For improving the apparatus and method of medical image quality
CN107533755B (en) * 2015-04-14 2021-10-08 皇家飞利浦有限公司 Apparatus and method for improving medical image quality

Also Published As

Publication number Publication date
US7321699B2 (en) 2008-01-22
AU2003270386A8 (en) 2004-03-29
AU2003270386A1 (en) 2004-03-29
US20040131273A1 (en) 2004-07-08
US20080095433A1 (en) 2008-04-24
WO2004023787A3 (en) 2004-05-13

Similar Documents

Publication Publication Date Title
US7321699B2 (en) Signal intensity range transformation apparatus and method
US8570396B2 (en) Multiple exposure high dynamic range image capture
US8525900B2 (en) Multiple exposure high dynamic range image capture
EP1058909B1 (en) A new perceptual thresholding for gradient-based local edge detection
EP2515273B1 (en) Multiple exposure high dynamic range image capture
JP2010525486A (en) Image segmentation and image enhancement
WO2003061266A2 (en) System and method for compressing the dynamic range of an image
US9165345B2 (en) Method and system for noise reduction in video systems
US11836888B2 (en) Method and apparatus for processing RGB-infrared (RGB-IR) sensor
US11551336B2 (en) Chrominance and luminance enhancing systems and methods
US9881358B2 (en) Method and system for adaptive pixel replacement
US20210287344A1 (en) Directional scaling systems and methods
US8897378B2 (en) Selective perceptual masking via scale separation in the spatial and temporal domains using intrinsic images for use in data compression
US10719916B2 (en) Statistical noise estimation systems and methods
GB2490231A (en) Multiple exposure High Dynamic Range image capture
US11321813B2 (en) Angular detection using sum of absolute difference statistics systems and methods
CN114972087A (en) Video processing method, device, equipment and computer storage medium
CN115731130A (en) Image processing method, image processing apparatus, imaging device, electronic device, and storage medium
Steux et al. of Deliverable: Report on Computer Vision Algorithms
JPH08265555A (en) Picture processor

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: JP

WWW Wipo information: withdrawn in national office

Country of ref document: JP