WO1991008648A1 - A hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment - Google Patents

A hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment Download PDF

Info

Publication number
WO1991008648A1
WO1991008648A1 PCT/US1990/006718 US9006718W WO9108648A1 WO 1991008648 A1 WO1991008648 A1 WO 1991008648A1 US 9006718 W US9006718 W US 9006718W WO 9108648 A1 WO9108648 A1 WO 9108648A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
residual
steps
resolution
images
Prior art date
Application number
PCT/US1990/006718
Other languages
French (fr)
Inventor
Paul William Jones
Paul Walter Melnychuck
Original Assignee
Eastman Kodak Company
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 Eastman Kodak Company filed Critical Eastman Kodak Company
Priority to EP91900444A priority Critical patent/EP0455794B1/en
Priority to DE69029725T priority patent/DE69029725T2/en
Publication of WO1991008648A1 publication Critical patent/WO1991008648A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53—Querying
    • G06F16/532—Query formulation, e.g. graphical querying
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N11/00—Colour television systems
    • H04N11/04—Colour television systems using pulse code modulation
    • H04N11/042—Codec means
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/59—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets

Definitions

  • the present invention is related to a method of storing and displaying digital images and more particularly to a multiresolution method which makes available reduced resolution versions of original images for quick display and full resolution original images for making photographic quality hardcopies.
  • a multiuse environment is one where the user has the option of selecting the type of display medium and the desired resolution on this medium.
  • two types of display media are considered: video monitors and color hardcopies, through photographic, thermal imaging, or other means. Versions of an original digital image are made available at various resolutions to allow for the display of an HDTV quality image on video, an NTSC quality image with PAL/SECAM compatibility on video, and a sub-NTSC quality image on video, as well as a very high q ality image on color hardcopy.
  • the techniques described herein are not limited to these display media or resolutions.
  • the present invention uses a multiresolution or hierarchical structuring of the image data to allow for rapid access to the various image formats. It is also desirable to encode this hierarchical image data in an efficient manner so as to reduce the storage requirements on a digital storage media, e.g., optical discs or magnetic tape, and to allow for the rapid transfer of the image data from the storage media to the output device. Furthermore, it is desirable to minimize the hardware required to implement such a system. In particular, it is desirable to minimize the hardware required to access the lower resolution images along with its cost. Systems which use a hierarchical structuring of the image data are often termed progressive transmission since they allow image quality to be progressively refined; however, they are
  • Laplacian pyramid, the Burt pyramid, or the residual pyramid In this technique, the original image G 0 is lowpass filtered (LPF) , and this lowpass image is subsampled to take advantage of its reduced bandwidth to provide the image G ]. , .
  • LPF lowpass filtered
  • This process of lowpass filtering and decimation is repeated three, times to generate a hierarchical structure, or pyramid, of images (G 0 , (? ⁇ , G , and G 3 ) of successively smaller dimensions. Although four resolution levels are shown, more or less can be used, depending on the application.
  • the residual images corresponding to the levels of the lowpass pyramid form another pyramid which is termed the Laplacian, Burt, or residual pyramid. This technique is motivated by the fact that the residual images have a reduced variance and entropy compared to the original or lowpass images and may be quantized and entropy encoded to provide efficient s ⁇ torage ⁇ of the data.
  • the encoded residuals are denoted 0 , L l t L 2 , and L- .
  • Reconstruction is performed by interpolating the decoded lowpass image at the bottom of the lowpass pyramid and adding in the corresponding decoded residual to generate the next level in the lowpass pyramid. This process is iterated until the original image size is reached; the decoding process is illustrated in Figure 2.
  • a progressive improvement in reconstructed image quality and resolution can thus be obtained by displaying the reconstructed lowpass filtered image at each level of the pyramid, i.e., G 3 ,
  • the patent to Anderson teaches a modification of the Burt pyramid scheme in which the lowpass filtered image is subtracted directly from the image at the previous level and then subsampled; it is not subsampled and interpolated prior to subtraction as in the Burt pyramid.
  • This technique is termed the filter-subtract-decimate (FSD) pyramid.
  • FSD filter-subtract-decimate
  • the patent to Bes ⁇ ler et al (4,709,394) and Carlson et al (4,674,125) teach specific real-time implementations of the Burt pyramid or the FSD pyramid.
  • the patent to Itoh et al (4,682,869) essentially teaches the method of the Burt pyramid, albeit for the specific case of the multiuse scenario.
  • the main advancement in the patent to Itoh et al seems to be specific methods for forming the prediction image used to generate the residual image and methods for encoding the residual, particularly for the case of binary images.
  • the techniques require quantization of the hierarchical image data to achieve significant compression. If the hierarchy contains many levels of resolution (as is required in the multiuse scenario) , significant degradation may occur at the highest resolution level owing to propagation of quantization noise. Therefore, the prior art techniques are undesirable. '
  • the to be disclosed system fills the need for very high quality at the highest resolution level (for hardcopies of photographic quality) while providing efficient access to images at lower resolution levels.
  • the present invention overcomes these limitations by using a hybrid scheme which permits fast reconstruction of the lower resolution images with relatively simple hardware while yielding very high reconstruction quality at the highest resolution level. Furthermore, the image hierarchy can still be encoded efficiently, i.e., at a low bit rate, and is stored in such a manner that it can be accessed efficiently when being read from the storage medium. In essence, the proposed technique uses the residual pyramid to provide efficient storage of the higher resolution levels combined with other techniques for the lower resolution level ⁇ to satisfy the system requirement of fast access to these lower levels.
  • the image data is encoded using an entropy coding technique such as Huffman coding or Q-coding, although it may be advantageous in some cases to store the data without entropy coding to provide immediate access to these records.
  • the data may also be quantized prior to encoding, or it may be encoded in a lossless manner, i.e., without quantization.
  • the present method effectively decouples the lower resolution images from tile higher resolution images to prevent the propagation of quantization noise from the lower levels and to provide quick access to the lower resolution images. From the foregoing, it can be seen that it is a primary object of the present invention to provide a. method for storing and displaying digital images with selectable multiresolution levels.
  • Figure 1 is a block diagram illustrating a prior art pyramid residual image encoder method.
  • Figure 2 is a block diagram illustrating a prior art pyramid residual image decoder method.
  • Figure 3 is a block diagram illustrating the hierarchical storage and display method.
  • Figure 4 is a block diagram illustrating the decomposition and encoder portion of a first preferred embodiment of the present invention using a residual pyramid with progressive partitioning.
  • Figure 5 is a block diagram illustrating the progressive partitioning of a full resolution image into lesser resolution images.
  • Figure 6 is a block diagram illustrating the decoder and reconstruction portion of a first preferred embodiment of the present invention using a residual pyramid with progressive partitioning.
  • Figure 7 is a block diagram illustrating the decomposition and encoder portion of a second preferred embodiment of the present invention using a residual pyramid with direct access to multiresolution images.
  • Figure 8 is a block diagram illustrating the decoder and reconstruction portion of a second preferred embodiment of the present invention using a residual pyramid with direct access to multiresolution images.
  • Figure 9 is a block diagram illustrating the decomposition and encoder portion of a third preferred embodiment of the present invention using a residual pyramid with CD-I compatibility.
  • Figure 10 is a block diagram illustrating the decoder and reconstruction portion of a third preferred embodiment of the present invention using a residual pyramid with CD-I compatibility.
  • Figure 11 is a, block diagram illustrating the color conversion method.
  • Figure 12 is a block diagram illustrating the decomposition and encoder portion of a first preferred embodiment of the present invention for chrominance signals.
  • Figure 13 is a ' block diagram illustrating the decoder and reconstruction portion of a first, preferred embodiment of the present invention for chrominance signals.
  • Figure 14 is a block diagram illustrating the decomposition and encoder portion of a second preferred embodiment of the present invention for chrominance signals.
  • Figure 15 is a block diagram illustrating the decoder and reconstruction portion of a second preferred embodiment of the present invention for chrominance signals.
  • Figure 16 is a block diagram illustrating the decomposition and encoder portion of a third preferred embodiment of the present invention for chrominance signals.
  • Figure 17 is a block diagram illustrating the decoder and reconstruction portion of a third preferred embodiment of the present invention for chrominance ' signals.
  • Figure 18 is a block diagram illustrating an alternate method for the decoder and reconstruction portion of a third preferred embodiment of the present invention for chrominance signals.
  • Figure 19 is a block diagram illustrating the decomposition and encoder portion of a fourth preferred embodiment of the present invention for chrominance signals.
  • Figure 20 is a block diagram illustrating the decoder and reconstruction portion of a fourth preferred embodiment of the present invention for chrominance signals.
  • Figure 21 is a block diagram illustrating the decomposition and encoder portion of a fifth preferred embodiment, of the present invention for chrominance signals.
  • Figure 22 is a block diagram illustrating the decoder and reconstruction portion of a fifth preferred embodiment of the present invention for chrominance signals.
  • Figure 23 is a block diagram illustrating an alternate method for the decoder and reconstruction portion of a fifth preferred embodiment of the present invention for chrominance signals.
  • the highest resolution image is composed of 2048 x 3072 pixels and that this resolution is adequate to produce photographic quality originals on an appropriate digital output device. It is also assumed that the lowest resolution level is composed of 256 x 384 pixels and that this resolution is adequate for preliminary, but relatively low quality, display on an NTSC-quality video monitor.
  • the other resolution levels are 512 x 768 pixels for high quality display on an NTSC-quality video monitor and 1024 x 1536 for high quality display on proposed HDTV video monitors.
  • resolutions will be denoted as 256 (for 256 x 384), 512 (for 512 x 768), IK (for 1024 x 1536), and 2K (for 2048 x 3072) for convenience.
  • the basic diagram for the hierarchical storage and display method is illustrated in Figure 3.
  • the invention is not limited to these resolutions or output devices. Indeed, many variations with other resolutions and output devices may be used, and these resolutions may not necessarily be related to each other by some power of two as the previously described resolutions are (assuming some form of interpolation is available.)
  • the number of resolution levels and _the type of decomposition at each level may vary to some extent, yet still be consistent with the described embodiments.
  • the embodiments are first described in general terms for a single channel image, i.e., one-color or luminance image, and then further embodiments are described for the case of three-channel color images.
  • the 2K image is decomposed into a residual pyramid consisting of a 512 base image, a IK residual, and a 2K residual.
  • a separate technique is used whereby the 512 image is decomposed into four 256 images by subsa pling the original 512 image in a staggered fashion as shown in Figure 5.
  • This technique progressive partitioning.
  • the information to be stored is then composed of the four 256 partitioned images (which can be used to reconstruct the 512 image) and the IK and 2K residuals.
  • the residuals can then be used to reconstruct the IK image and the 2K image.
  • the reconstruction process is illustrated in Figure 6. This technique is motivated by the following:
  • the 256 partitioned images can be created and displayed without any special/additional hardware.
  • the 512 image can be built up and displayed in a progressive manner without any special hardware using one or more of the 256 partitioned images. Also, since the 256 images are direct ⁇ ubsamples of the 512 image, the reconstruction of the 512 image can be done without the addition of any quantization noise in excess of what it originally contained.
  • This technique is efficient in terms of nonredundant storage of the image data, and the use of the residuals still allows for significant image compression.
  • the 2K image is decomposed into a residual pyramid consisting of a 512 base image, a IK residual, and a 2K residual.
  • the 512 image is not decomposed into 256 images using progressive partitioning, but is merely prefiltered and subsampled to create a 256 image suitable for display.
  • This process is shown in Figure 7.
  • the information to be stored consists of the 256 image, the 512 image, the IK residual, and the 2K residual.
  • the residuals can be used to reconstruct the IK and 2K images.
  • the reconstruction process is illustrated in Figure 8. The advantages of this approach are:
  • the 256 and 512 images are directly available for display with no special hardware or filtering operations, while the use of residuals for the other resolution images still allows for significant image compression.
  • the lower resolution images are more completely decoupled from the 2K image so that very high fidelity can be achieved at the 2K level.
  • the 512 image is prefiltered prior to subsampling which allows for higher quality in the 256 image as compared to progressive partitioning.
  • the third embodiment uses a residual pyramid, but also incorporates the Philips CD-I (Compact Disc-Interactive) real-time video data format (as referenced in the publication by A. C. Luther in Prior Art) at the 256 and 512 levels in order to provide compatibility with this format.
  • the 2K image is decomposed into a residual pyramid consisting of a 512 image, a IK residual, and a 2K residual.
  • the 512 image is then lowpass filtered and subsampled to form a 256 base image as in the previous embodiment.
  • the technique differs in the encoding of the 256 base image and the 512 image in order to be compatible with the CD-I format.
  • the 256 base image is encoded using the CD-I encoding scheme, namely, differential pulse ' code modulation (DPCM) , rather than an entropy coding technique as was used in the other embodiments.
  • This 256 image is then decoded with the CD-I decoder and interpolated to the dimensions of the 512 image, forming a 512 prediction image.
  • a 512 residual image is formed by subtracting this prediction image from the 512 image.
  • the 512 residual image is first encoded using the CD-I "extended resolution" format which is a 7-bit quantization of the residual.
  • a secondary 512 residual is then formed as the difference between the original 512 residual and the 7-bit CD-I residual.
  • the information to be stored thus consists of the CD-I 256 base image, the CD-I 512 extended resolution residual, the 512 secondary residual, the IK residual, and the 2K residual images.
  • the reconstruction decoding process is illustrated in Figure 10. The advantage of this technique is the compatibility with the CD-I format at the 256 and 512 levels.
  • the described embodiments may be extended to handle images that are composed of more than one-channel, specifically, three- channel color images.
  • the direct approach is to encode each channel in the same manner, using one of the described embodiments.
  • the three-color signal is comprised of a luminance signal, Y, and two chrominance signals, C ] _, and C 2 . These signals may be acquired directly from devices designed to output such color signals, or they may be generated by converting red
  • B signals and hence an inverse color conversion from Y, Ci , and C 2 to R, G, and B is also a linear transformation.
  • the color conversion apparatus is illustrated in Figure 11.
  • An example of the luminance/chrominance conversion from R, G, and B signals is the NTSC (National Television Standards Committee) YIQ conversion given by the following equations.
  • Y 0.299R + 0.587G + 0.114B
  • the chrominance signals C l t and C 2 may also be simple color difference signals such as R - Y and
  • the human visual system has a reduced bandwidth for the chrominance signals as compared to the luminance signal, and thus a substantial savings in storage can be achieved by subsampling the chrominance signals, C 1( and C 2 relative to the luminance signal Y.
  • Typical subsampling factors for C 1 and C 2 relative to Y at a given resolution level are 2 or 4. It may be advantageous, to prefilter the chrominance signals " prior to subsampling reduce aliasing artifacts.
  • the luminance signal is encoded using one of the embodiments described previously.
  • any of the preferred embodiments for chrominance signals may be modified by using DPCM encoding and decoding of the 256 base image.
  • the apparatus of Figure 12 is used for decomposition and encoding.
  • the chrominance information to be stored consists solely of a 256 base image for each chrominance signal.
  • the apparatus of Figure 13 is used.
  • This embodiment has the advantage of requiring a minimal amount of storage space for the chrominance signals and requiring only an interpolation process to generate the chrominance signals at higher resolution levels.
  • a limitation of this method is that chrominance errors may be introduced at the higher resolution levels.
  • the apparatus of Figure 14 is used for decomposition and encoding.
  • the chrominance information to be stored consists of a 256 base image and a 512 residual. Images are reconstructed using the apparatus of Figure 15. This method improves reconstructed quality at the 512, IK, and 2K levels compared to the first embodiment while still providing significant compression of the chrominance signals.
  • the apparatus of Figure 16 is used for decomposition and encoding.
  • the chrominance information to be stored consists of a 256 base image and a IK residual.
  • Images can be reconstructed using the apparatus of Figure 17 or by using the apparatus of Figure 18 where the 512 chrominance signal is generated by subsampling the reconstructed IK chrominance signal. This method improves reconstructed quality at the IK and 2K levels compared to the second embodiment.
  • the apparatus of Figure 19 is used for decomposition and encoding.
  • the chrominance information to be stored consists of a 256 base image, a 512 residual image, and a IK residual image. Images are reconstructed using the apparatus of Figure 20.
  • the advantage of this embodiment over the third embodiment is that it is not necessary to reconstruct the IK image first before the 512 image can be computed (since the 512 image is reconstructed by adding a 512 residual to an interpolated 256 image.)
  • the disadvantage is the additional storage required as compared to the third embodiment.
  • the apparatus of Figure 21 is used for decomposition and encoding.
  • the chrominance information to be stored consists of a 256 base image and a 2K residual. Images are reconstructed using the apparatus of Figure 22 or by using the apparatus of Figure 23 where the 512 and IK chrominance signals are generated by subsampling the 2K reconstructed signal.
  • This method provides very high quality at the 2K level compared to the other embodiments, and still provides good quality at the 512 and IK levels since the chrominance signals at these levels can be generated from a higher resolution signal. Since residuals for the 512 and IK levels are not stored, this approach provides some compression over a full residual pyramid approach.

Abstract

An image processing method is described for the hierarchical storage and display of high resolution digital images in a multiuse environment. The method makes reduced resolution versions of the original image available to quick display on video monitors while providing access to the full resolution image for photographic quality hardcopies. This multiresolution method also provides for the efficient storage of this data via a hybrid coding scheme based on residuals. Several embodiments of the method are described.

Description

A HYBRID RESIDUAL-BASED HIERARCHICAL STORAGE
AND DISPLAY METHOD FOR HIGH RESOLUTION DIGITAL
IMAGES IN A MULTIUSE ENVIRONMENT
TECHNICAL FIELD OF THE INVENTION
The present invention is related to a method of storing and displaying digital images and more particularly to a multiresolution method which makes available reduced resolution versions of original images for quick display and full resolution original images for making photographic quality hardcopies.
BACKGROUND OF THE INVENTION
An image processing method will be described for the efficient storage and display of high resolution digital images in a ultiuse environment. A multiuse environment is one where the user has the option of selecting the type of display medium and the desired resolution on this medium. In particular, two types of display media are considered: video monitors and color hardcopies, through photographic, thermal imaging, or other means. Versions of an original digital image are made available at various resolutions to allow for the display of an HDTV quality image on video, an NTSC quality image with PAL/SECAM compatibility on video, and a sub-NTSC quality image on video, as well as a very high q ality image on color hardcopy. However, it is important to note that the techniques described herein are not limited to these display media or resolutions.
The present invention uses a multiresolution or hierarchical structuring of the image data to allow for rapid access to the various image formats. It is also desirable to encode this hierarchical image data in an efficient manner so as to reduce the storage requirements on a digital storage media, e.g., optical discs or magnetic tape, and to allow for the rapid transfer of the image data from the storage media to the output device. Furthermore, it is desirable to minimize the hardware required to implement such a system. In particular, it is desirable to minimize the hardware required to access the lower resolution images along with its cost. Systems which use a hierarchical structuring of the image data are often termed progressive transmission since they allow image quality to be progressively refined; however, they are
"generally not designed specifically for a multiuse environment.
Prior Art A number of hierarchical structures for image- coding and/or progressive transmission have been described in the open technical literature and in . various patents. Of particular relevance to the present invention are the following publications:
P. J. Burt and E. H. Adelson,
"The Laplacian Pyramid as a Compact Image Code, "
IEEE Trans. Co mun. , COM-31, 532-540 (1983) .
• A. C. Luther,
"Digital Video in the. PC Environment, " McGraw-Hill Book Company, New York, 81-84 (1989).
and the following patents: .
U.S. Patent No. 4,709,394, entitled "Multiplexed Real-Time Pyramid Signal Processing System", by Bessler et al., 11/24/87.
• U.S. Patent No. 4,674,125, entitled "Real-Time
Hierarchical Pyramid Signal Processing Apparatus", by Carlson et al., 6/16/87.
U.S. Patent No. 4,718,104, entitled "Filter- Subtract-Decimate Hierarchical Pyramid Signal Analyzing and Synthesizing Technique", by Anderson, 1/5/88.
U.S. Patent No. 4,682,869, entitled "Image Processing System and Method", by Itoh et al., 7/28/87.
Referring to Figure 1, the publication by Burt et al, teaches an encoding method for images termed the
Laplacian pyramid, the Burt pyramid, or the residual pyramid. In this technique, the original image G0 is lowpass filtered (LPF) , and this lowpass image is subsampled to take advantage of its reduced bandwidth to provide the image G]., . This process of lowpass filtering and decimation is repeated three, times to generate a hierarchical structure, or pyramid, of images (G0, (?ι, G , and G3) of successively smaller dimensions. Although four resolution levels are shown, more or less can be used, depending on the application.
Each lowpass image in this pyramid is then expanded to the dimensions of the next higher level by upsa pling
(inserting zeros) and filtering to form a prediction image for that level. This prediction image is subtracted from its corresponding lowpass image in a subtractor to generate difference, or residual, images LQ , Lλ , and L2. At the final level, the residual image is defined to be equivalent to the lowpass image, i.e., 3 = G3. The residual images corresponding to the levels of the lowpass pyramid form another pyramid which is termed the Laplacian, Burt, or residual pyramid. This technique is motivated by the fact that the residual images have a reduced variance and entropy compared to the original or lowpass images and may be quantized and entropy encoded to provide efficient s Λtorage Λof the data. The encoded residuals are denoted 0, Ll t L2 , and L- . Reconstruction is performed by interpolating the decoded lowpass image at the bottom of the lowpass pyramid and adding in the corresponding decoded residual to generate the next level in the lowpass pyramid. This process is iterated until the original image size is reached; the decoding process is illustrated in Figure 2. A progressive improvement in reconstructed image quality and resolution can thus be obtained by displaying the reconstructed lowpass filtered image at each level of the pyramid, i.e., G3,
Λ Λ
G2, G- , and G0. Note that errors introduced in the encoding process are propagated from one level to the next higher level in the decoding process.
The patent to Anderson (4,718,104) teaches a modification of the Burt pyramid scheme in which the lowpass filtered image is subtracted directly from the image at the previous level and then subsampled; it is not subsampled and interpolated prior to subtraction as in the Burt pyramid. This technique is termed the filter-subtract-decimate (FSD) pyramid. The primary advantage of this technique is a reduction in the hardware required to generate the pyramid.
The patent to Besεler et al (4,709,394) and Carlson et al (4,674,125) teach specific real-time implementations of the Burt pyramid or the FSD pyramid. The patent to Itoh et al (4,682,869) essentially teaches the method of the Burt pyramid, albeit for the specific case of the multiuse scenario. The main advancement in the patent to Itoh et al seems to be specific methods for forming the prediction image used to generate the residual image and methods for encoding the residual, particularly for the case of binary images. There are two primary limitations in using the techniques described in the prior art for a multiuse environment:
• To reconstruct the lowpass signal at a given resolution level, all lower resolution images must be reconstructed first. This requires substantial hardware and processing time.
• The techniques require quantization of the hierarchical image data to achieve significant compression. If the hierarchy contains many levels of resolution (as is required in the multiuse scenario) , significant degradation may occur at the highest resolution level owing to propagation of quantization noise. Therefore, the prior art techniques are undesirable.'
In contrast, the to be disclosed system fills the need for very high quality at the highest resolution level (for hardcopies of photographic quality) while providing efficient access to images at lower resolution levels.
SUMMARY OF THE INVENTION The present invention overcomes these limitations by using a hybrid scheme which permits fast reconstruction of the lower resolution images with relatively simple hardware while yielding very high reconstruction quality at the highest resolution level. Furthermore, the image hierarchy can still be encoded efficiently, i.e., at a low bit rate, and is stored in such a manner that it can be accessed efficiently when being read from the storage medium. In essence, the proposed technique uses the residual pyramid to provide efficient storage of the higher resolution levels combined with other techniques for the lower resolution levelε to satisfy the system requirement of fast access to these lower levels. It is assumed the image data is encoded using an entropy coding technique such as Huffman coding or Q-coding, although it may be advantageous in some cases to store the data without entropy coding to provide immediate access to these records. The data may also be quantized prior to encoding, or it may be encoded in a lossless manner, i.e., without quantization. To some extent, the present method effectively decouples the lower resolution images from tile higher resolution images to prevent the propagation of quantization noise from the lower levels and to provide quick access to the lower resolution images. From the foregoing, it can be seen that it is a primary object of the present invention to provide a. method for storing and displaying digital images with selectable multiresolution levels.
It is another object of the present invention to provide a hierarchical storage and display method which minimizes hardware requirements, storage space, and cost.
The above and other objects of the present invention will become more apparent when taken in conjunction with the following description and drawings wherein like characters indicate like parts and which drawings form a part of the present description.
BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram illustrating a prior art pyramid residual image encoder method.
Figure 2 is a block diagram illustrating a prior art pyramid residual image decoder method.
Figure 3 is a block diagram illustrating the hierarchical storage and display method. Figure 4 is a block diagram illustrating the decomposition and encoder portion of a first preferred embodiment of the present invention using a residual pyramid with progressive partitioning. Figure 5 is a block diagram illustrating the progressive partitioning of a full resolution image into lesser resolution images.
Figure 6 is a block diagram illustrating the decoder and reconstruction portion of a first preferred embodiment of the present invention using a residual pyramid with progressive partitioning.
Figure 7 is a block diagram illustrating the decomposition and encoder portion of a second preferred embodiment of the present invention using a residual pyramid with direct access to multiresolution images. Figure 8 is a block diagram illustrating the decoder and reconstruction portion of a second preferred embodiment of the present invention using a residual pyramid with direct access to multiresolution images.
Figure 9 is a block diagram illustrating the decomposition and encoder portion of a third preferred embodiment of the present invention using a residual pyramid with CD-I compatibility. Figure 10 is a block diagram illustrating the decoder and reconstruction portion of a third preferred embodiment of the present invention using a residual pyramid with CD-I compatibility.
Figure 11 is a, block diagram illustrating the color conversion method.
Figure 12 is a block diagram illustrating the decomposition and encoder portion of a first preferred embodiment of the present invention for chrominance signals. Figure 13 is a' block diagram illustrating the decoder and reconstruction portion of a first, preferred embodiment of the present invention for chrominance signals.
Figure 14 is a block diagram illustrating the decomposition and encoder portion of a second preferred embodiment of the present invention for chrominance signals.
Figure 15 is a block diagram illustrating the decoder and reconstruction portion of a second preferred embodiment of the present invention for chrominance signals.
Figure 16 is a block diagram illustrating the decomposition and encoder portion of a third preferred embodiment of the present invention for chrominance signals. Figure 17 is a block diagram illustrating the decoder and reconstruction portion of a third preferred embodiment of the present invention for chrominance' signals.
Figure 18 is a block diagram illustrating an alternate method for the decoder and reconstruction portion of a third preferred embodiment of the present invention for chrominance signals.
Figure 19 is a block diagram illustrating the decomposition and encoder portion of a fourth preferred embodiment of the present invention for chrominance signals.
Figure 20 is a block diagram illustrating the decoder and reconstruction portion of a fourth preferred embodiment of the present invention for chrominance signals.
Figure 21 is a block diagram illustrating the decomposition and encoder portion of a fifth preferred embodiment, of the present invention for chrominance signals. Figure 22 is a block diagram illustrating the decoder and reconstruction portion of a fifth preferred embodiment of the present invention for chrominance signals.
Figure 23 is a block diagram illustrating an alternate method for the decoder and reconstruction portion of a fifth preferred embodiment of the present invention for chrominance signals.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
In the following description of the preferred embodiments, it will be assumed that the highest resolution image is composed of 2048 x 3072 pixels and that this resolution is adequate to produce photographic quality originals on an appropriate digital output device. It is also assumed that the lowest resolution level is composed of 256 x 384 pixels and that this resolution is adequate for preliminary, but relatively low quality, display on an NTSC-quality video monitor. The other resolution levels are 512 x 768 pixels for high quality display on an NTSC-quality video monitor and 1024 x 1536 for high quality display on proposed HDTV video monitors. These resolutions will be denoted as 256 (for 256 x 384), 512 (for 512 x 768), IK (for 1024 x 1536), and 2K (for 2048 x 3072) for convenience. The basic diagram for the hierarchical storage and display method is illustrated in Figure 3. However, it is important to note that the invention is not limited to these resolutions or output devices. Indeed, many variations with other resolutions and output devices may be used, and these resolutions may not necessarily be related to each other by some power of two as the previously described resolutions are (assuming some form of interpolation is available.) In addition, the number of resolution levels and _the type of decomposition at each level may vary to some extent, yet still be consistent with the described embodiments. The embodiments are first described in general terms for a single channel image, i.e., one-color or luminance image, and then further embodiments are described for the case of three-channel color images.
1. Residual Pyramid With Progressive Partitioning
In the first embodiment, illustrated in Figure 4, the 2K image is decomposed into a residual pyramid consisting of a 512 base image, a IK residual, and a 2K residual. To further decompose the 512 image, a separate technique is used whereby the 512 image is decomposed into four 256 images by subsa pling the original 512 image in a staggered fashion as shown in Figure 5. We term this technique progressive partitioning. The information to be stored is then composed of the four 256 partitioned images (which can be used to reconstruct the 512 image) and the IK and 2K residuals. The residuals can then be used to reconstruct the IK image and the 2K image. The reconstruction process is illustrated in Figure 6. This technique is motivated by the following:
The 256 partitioned images can be created and displayed without any special/additional hardware.
The 512 image can be built up and displayed in a progressive manner without any special hardware using one or more of the 256 partitioned images. Also, since the 256 images are direct εubsamples of the 512 image, the reconstruction of the 512 image can be done without the addition of any quantization noise in excess of what it originally contained.
• The lowest, resolution image is effectively decoupled from the higher resolution images to minimize the propagation of quantization noise to the higher levels.
This technique is efficient in terms of nonredundant storage of the image data, and the use of the residuals still allows for significant image compression.
2. Residual Coding With Direct Access To 256 And 512 Images
In the second embodiment, direct access is provided to both the 256 and 512 images, i.e., no progressive partitioning is used. In this approach, the 2K image is decomposed into a residual pyramid consisting of a 512 base image, a IK residual, and a 2K residual. The 512 image is not decomposed into 256 images using progressive partitioning, but is merely prefiltered and subsampled to create a 256 image suitable for display. This process is shown in Figure 7. The information to be stored consists of the 256 image, the 512 image, the IK residual, and the 2K residual. The residuals can be used to reconstruct the IK and 2K images. The reconstruction process is illustrated in Figure 8.The advantages of this approach are:
The 256 and 512 images are directly available for display with no special hardware or filtering operations, while the use of residuals for the other resolution images still allows for significant image compression.
The lower resolution images are more completely decoupled from the 2K image so that very high fidelity can be achieved at the 2K level. The 512 image is prefiltered prior to subsampling which allows for higher quality in the 256 image as compared to progressive partitioning.
The tradeoff with this approach is that additional storage is required since the 256 image is redundant to the 512 image.
3. Residual Pyramid With CD-I Compatibility Referring to Figure 9, the third embodiment uses a residual pyramid, but also incorporates the Philips CD-I (Compact Disc-Interactive) real-time video data format (as referenced in the publication by A. C. Luther in Prior Art) at the 256 and 512 levels in order to provide compatibility with this format. The 2K image is decomposed into a residual pyramid consisting of a 512 image, a IK residual, and a 2K residual. The 512 image is then lowpass filtered and subsampled to form a 256 base image as in the previous embodiment. However, the technique differs in the encoding of the 256 base image and the 512 image in order to be compatible with the CD-I format. The 256 base image is encoded using the CD-I encoding scheme, namely, differential pulse 'code modulation (DPCM) , rather than an entropy coding technique as was used in the other embodiments. This 256 image is then decoded with the CD-I decoder and interpolated to the dimensions of the 512 image, forming a 512 prediction image. A 512 residual image is formed by subtracting this prediction image from the 512 image. The 512 residual image is first encoded using the CD-I "extended resolution" format which is a 7-bit quantization of the residual. A secondary 512 residual is then formed as the difference between the original 512 residual and the 7-bit CD-I residual. The information to be stored thus consists of the CD-I 256 base image, the CD-I 512 extended resolution residual, the 512 secondary residual, the IK residual, and the 2K residual images. The reconstruction decoding process is illustrated in Figure 10. The advantage of this technique is the compatibility with the CD-I format at the 256 and 512 levels.
Of substantial interest, is how the described embodiments may be extended to handle images that are composed of more than one-channel, specifically, three- channel color images. The direct approach is to encode each channel in the same manner, using one of the described embodiments. However, it is possible with some color spaces to achieve a substantial savings in storage requirements by subsampling the data to exploit the reduced bandwidth of the human visual system in certain color channels. Thus, it may be advantageous to encode each color channel in a different manner. In' describing the apparatus used to encode such color signals, we assume that the three-color signal is comprised of a luminance signal, Y, and two chrominance signals, C]_, and C2. These signals may be acquired directly from devices designed to output such color signals, or they may be generated by converting red
(R) , green (G) , and blue (B) color signals supplied by a color television camera. The color conversion to Y, Ci , and C2 is a linear transformation of the R, G, and
B signals, and hence an inverse color conversion from Y, Ci , and C2 to R, G, and B is also a linear transformation. The color conversion apparatus is illustrated in Figure 11. An example of the luminance/chrominance conversion from R, G, and B signals is the NTSC (National Television Standards Committee) YIQ conversion given by the following equations. Y = 0.299R + 0.587G + 0.114B I = Cx = 0.596R - 0.273G - 0.322B
Q = C2 = 0.212R - 0.522G + 0.315B
The chrominance signals Cl t and C2 may also be simple color difference signals such as R - Y and
B - Y, for example. We define the color conversion to be of these forms or their like.
It is well-known that the human visual system has a reduced bandwidth for the chrominance signals as compared to the luminance signal, and thus a substantial savings in storage can be achieved by subsampling the chrominance signals, C1( and C2 relative to the luminance signal Y. Typical subsampling factors for C1 and C2 relative to Y at a given resolution level are 2 or 4. It may be advantageous, to prefilter the chrominance signals" prior to subsampling reduce aliasing artifacts. In the following discussion and drawings, it is assumed that the luminance signal is encoded using one of the embodiments described previously. Also, the embodiments are shown for only one chrominance signal, and it is understood that C^, and C2 may be encoded using the same method or possibly differing methods to take advantage of each signal's specific characteristics. To provide compatibility with the Philips CD-I format, any of the preferred embodiments for chrominance signals may be modified by using DPCM encoding and decoding of the 256 base image. In a first embodiment for chrominance signals, the apparatus of Figure 12 is used for decomposition and encoding. The chrominance information to be stored consists solely of a 256 base image for each chrominance signal. To reconstruct images at the various resolution levels, the apparatus of Figure 13 is used. This embodiment has the advantage of requiring a minimal amount of storage space for the chrominance signals and requiring only an interpolation process to generate the chrominance signals at higher resolution levels. A limitation of this method is that chrominance errors may be introduced at the higher resolution levels.
In a second embodiment for chrominance signals, the apparatus of Figure 14 is used for decomposition and encoding. The chrominance information to be stored consists of a 256 base image and a 512 residual. Images are reconstructed using the apparatus of Figure 15. This method improves reconstructed quality at the 512, IK, and 2K levels compared to the first embodiment while still providing significant compression of the chrominance signals. In a third embodiment for chrominance signals, the apparatus of Figure 16 is used for decomposition and encoding. The chrominance information to be stored consists of a 256 base image and a IK residual. Images can be reconstructed using the apparatus of Figure 17 or by using the apparatus of Figure 18 where the 512 chrominance signal is generated by subsampling the reconstructed IK chrominance signal. This method improves reconstructed quality at the IK and 2K levels compared to the second embodiment. In a fourth embodiment for chrominance signals, the apparatus of Figure 19 is used for decomposition and encoding. The chrominance information to be stored consists of a 256 base image, a 512 residual image, and a IK residual image. Images are reconstructed using the apparatus of Figure 20. The advantage of this embodiment over the third embodiment is that it is not necessary to reconstruct the IK image first before the 512 image can be computed (since the 512 image is reconstructed by adding a 512 residual to an interpolated 256 image.) The disadvantage is the additional storage required as compared to the third embodiment.
In a fifth embodiment for chrominance signals, the apparatus of Figure 21 is used for decomposition and encoding. The chrominance information to be stored consists of a 256 base image and a 2K residual. Images are reconstructed using the apparatus of Figure 22 or by using the apparatus of Figure 23 where the 512 and IK chrominance signals are generated by subsampling the 2K reconstructed signal. This method provides very high quality at the 2K level compared to the other embodiments, and still provides good quality at the 512 and IK levels since the chrominance signals at these levels can be generated from a higher resolution signal. Since residuals for the 512 and IK levels are not stored, this approach provides some compression over a full residual pyramid approach.
While there has been shown what are considered to be the preferred embodiments of the invention, it will be manifest that many changes and modifications may be made therein without departing from the essential spirit of the invention. It is intended, therefore, in the annexed claims, to cover all such changes and modifications as may fall within the true scope of the invention.

Claims

CLAIMS :
1. A hierarchical storage method comprising the steps of: a) forming a digital representation of an image; b) decomposing said digital representation according to the following steps:
1. lowpass filtering and subsampling said digital representation to form a lower resolution image;
2. interpolating or replicating said lower resolution image to the dimensions of the higher resolution image to form a prediction image; 3. subtracting said prediction image from the higher resolution image to form a residual image;
4. encoding and storing said residual image; and 5. creating a plurality of encoded and stored residual images of different resolutions by iterating steps 1, 2, 3, and 4 where the input to the lowpass filtering and subsampling operation of step 1 is the lower resolution image formed in the previous iteration of step 1; c) further decomposing the lowest
■ resolution image formed in b) into a plurality of lower resolution images according to the following steps: subsampling the even-numbered pixels on the even-numbered lines to create the first partitioned image; subsampling the odd-numbered pixels on the even-numbered lines to create the first partitioned image; subsampling the even-numbered pixels on the odd-numbered lines to create the first partitioned image; and subsampling the odd-numbered pixels on the odd-numbered lines to create the first partitioned image; d) encoding and storing said partitioned images.
2. A hierarchical display method according to Claim 1 and further comprising the steps of: retrieving and decoding one of said lower resolution images of c) ; and displaying said one retrieved image.
3. A hierarchical display method according to Claim 2 and further comprising the steps of: " retrieving and decoding at least one of the remainder of said lower resolution images of c) ,* combining all of said retrieved lower resolution images to form a higher resolution image for display; and displaying said combined retrieved image.
4. A hierarchical display method according to Claim 3 and further, comprising the steps of: retrieving and decoding all of said lower resolution images of c) ; combining all of said retrieved lower resolution images to form a higher resolution image for display; and displaying said combined retrieved image.
5. A hierarchical display method according to Claim 4 and further comprising the steps of: interpolating or replicating said combined retrieved image to the dimensions of the next higher resolution level to form a prediction image; retrieving and decoding the residual image corresponding to said next higher resolution level; adding said residual image to said prediction image to form a higher resolution image for display; displaying said higher resolution image; and iterating these steps for each remaining residual image of b) .
6. A hierarchical storage method comprising the steps of: a) forming a digital representation of an image; b) decomposing said digital representation according to the following steps:
1. lowpass filtering and subsampling said digital representation to form a lower resolution image;
2. interpolating or replicating said lower resolution image to the dimensions of the higher resolution image to form a prediction image;
3. subtracting said prediction image from the higher resolution image to form a residual image;
4. encoding and storing said residual image; and
5. creating a plurality of encoded and stored residual images of different resolutions by iterating steps 1, 2, 3, and 4 where the input to the lowpass filtering and subsampling operation of step 1 is the lower resolution image formed in the previous iteration of step 1; c) further decomposing the lowest resolution image formed in b) into a plurality of lower resolution images according to the following steps: iteratively lowpass filtering and subsampling said image to create one or more lower resolution images; and encoding and storing said lower resolution images.
7. A hierarchical display method according to Claim 6 and further comprising the steps of: retrieving and decoding one of said lower resolution images of c) ; and displaying said retrieved image.
8. A hierarchical display method according to Claim 7 and further comprising the steps of: retrieving and decoding said lowest resolution image of b) ; and displaying said retrieved image.
9. A hierarchical display method according to Claim 8 and further comprising the steps of: interpolating or replicating said retrieved image to the dimensions of the next higher resolution level to form a prediction image; retrieving and decoding the residual image corresponding to said next higher resolution level; adding said residual image to said prediction image to form a higher resolution image for display; displaying said higher resolution image; and iterating these steps for each remaining residual image of b) .
10. A hierarchical storage method comprising the steps of; a) forming a digital representation of an image; b) decomposing said digital representation according to the following steps:
1. lowpass filtering and subsampling said digital representation to form a lower resolution image; 2. interpolating or replicating said lower resolution image to the dimensions of the higher resolution image to form a prediction image;
3. subtracting said prediction image from the higher resolution image to form a residual- image;
4. encoding and storing said .residual image; and
5. creating a plurality of encoded and stored residual images of 'different resolutions by iterating steps 1, 2, 3, and 4 where the input to the lowpass filtering and subsampling operation of step 1 is the lower resolution image formed in the previous iteration of step 1; c) further decomposing the lowest resolution image formed in b) into one or more lower resolution images by iteratively lowpass filtering and subsampling the image until the lowest resolution image formed has dimensions of 256 pixels x 384 lines; d) encoding and storing the lowest resolution image of c) using the Philips CD-I (Compact Disc-Interactive) real-time video data format; and e) further decomposing the remaining images of c) according to the following steps:
1. decoding the lowest resolution image of d) ; 2. interpolating or replicating said decoded image to the dimensions of the next higher resolution level to form a prediction image;
3. subtracting said prediction image from the next higher resolution image formed in c) to create a residual image;
4. quantizing said residual image using the Philips CD-I 7-level quantization scheme to form a CD-I extended resolution residual; 5. encoding and storing said CD-I extended resolution residual using the Philips CD-I real-time video format;
6. subtracting said CD-I extended resolution residual from the residual of step 3 to form a secondary residual;
7. encoding and storing said residual;
8. iterating steps 2, 3, and 7 for each remaining image in c) .
11. A hierarchical display method according to Claim 10 and further comprising the steps of: retrieving and decoding said lowest resolution image of c) ; and displaying said retrieved image.
12. A hierarchical display method according to Claim 11 and further comprising the steps of: interpolating or replicating said retrieved image to the dimensions of the next higher resolution level to form a prediction image, retrieving and decoding the CD-I extended resolution residual of e) ; adding said prediction image to said decoded extended resolution residual to reconstruct an image for display; and displaying said reconstructed image.
13. A hierarchical display method according to Claim 12 and further comprising the steps of: retrieving and decoding the secondary residual of e) ; adding said reconstructed image to said decoded secondary residual to reconstruct an image for display; and displaying said reconstructed image.
14. A hierarchical display method according to Claim 13 and further comprising the steps of: interpolating or replicating said reconstructed image to the dimensions of the next higher resolution level to form a prediction image; retrieving and decoding the corresponding residual of e) ; adding said reconstructed image to said decoded residual to reconstruct an image for display; displaying said reconstructed image; and iterating these steps until the lowest resolution image of b) is reconstructed.
15. A hierarchical display method according to Claim 14 and further comprising the steps of: interpolating or replicating said reconstructed image to the dimensions of the next higher resolution level to form a prediction image; retrieving and decoding the corresponding residual of b) ; adding said reconstructed image to said decoded residual to reconstruct an image for display; displaying said reconstructed image; and iterating these steps for each remaining residual of b) .
PCT/US1990/006718 1989-11-29 1990-11-19 A hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment WO1991008648A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP91900444A EP0455794B1 (en) 1989-11-29 1990-11-19 A hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment
DE69029725T DE69029725T2 (en) 1989-11-29 1990-11-19 HYBRID HIERARCHICAL PROCESS BASED ON REMAINS, FOR STORING AND DISPLAYING HIGH-RESOLUTION DIGITAL IMAGES IN A MULTIPURPOSE ENVIRONMENT

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US07/442,872 US4969204A (en) 1989-11-29 1989-11-29 Hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment
US442,872 1989-11-29

Publications (1)

Publication Number Publication Date
WO1991008648A1 true WO1991008648A1 (en) 1991-06-13

Family

ID=23758486

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1990/006718 WO1991008648A1 (en) 1989-11-29 1990-11-19 A hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment

Country Status (5)

Country Link
US (1) US4969204A (en)
EP (1) EP0455794B1 (en)
JP (1) JPH04503295A (en)
DE (1) DE69029725T2 (en)
WO (1) WO1991008648A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6370274B1 (en) 1993-10-14 2002-04-09 U.S. Philips Corporation Method of storing in coded form image information of moving images, diagnostic system using the method, and image coding and recording unit and image-information retrieval unit for use in such a system

Families Citing this family (105)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5136396A (en) * 1988-08-30 1992-08-04 Canon Kabushiki Kaisha Image encoding method
JP2584840B2 (en) * 1988-08-30 1997-02-26 キヤノン株式会社 Image hierarchical coding device
US5210623A (en) * 1989-12-21 1993-05-11 Eastman Kodak Company Apparatus and method for quantizing and/or reconstructing multi-dimensional digital image signals
NL9000424A (en) * 1990-02-22 1991-09-16 Philips Nv TRANSFER SYSTEM FOR DIGITALIZED TELEVISION IMAGES.
US5255105A (en) * 1990-03-17 1993-10-19 International Computers Limited Encoding image data
JP3013089B2 (en) * 1990-05-24 2000-02-28 富士通株式会社 Natural picture expression method
NL9002110A (en) * 1990-09-19 1992-04-16 Koninkl Philips Electronics Nv IMAGE SEARCH AND DISPLAY SYSTEM AND A REGISTRATION BEARING CODED IMAGES, METHOD FOR RECORDING ENCRYPTED IMAGES AND A SEARCH AND DISPLAY DEVICE.
US5440401A (en) * 1990-09-14 1995-08-08 Eastman Kodak Company Image database incorporating low resolution index image data
JP3372949B2 (en) * 1990-09-19 2003-02-04 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ Display image display control mechanism
US5373375A (en) * 1990-12-21 1994-12-13 Eastman Kodak Company Metric conversion mechanism for digital images in a hierarchical, multi-resolution, multi-use environment
US5852674A (en) * 1991-04-08 1998-12-22 Canon Kabushiki Kaisha Apparatus and method for processing images using position-dependent weighing factor
US5263136A (en) * 1991-04-30 1993-11-16 Optigraphics Corporation System for managing tiled images using multiple resolutions
EP0516477B1 (en) * 1991-05-30 1998-01-07 Canon Kabushiki Kaisha Compression enhancement in graphics systems
US5297219A (en) * 1991-06-27 1994-03-22 Eastman Kodak Company Transforms for digital images in a hierarchical environment
US5227789A (en) * 1991-09-30 1993-07-13 Eastman Kodak Company Modified huffman encode/decode system with simplified decoding for imaging systems
JP2865467B2 (en) * 1991-11-22 1999-03-08 富士写真フイルム株式会社 Image data supply system
US5434838A (en) * 1991-12-05 1995-07-18 Fuji Photo Film Co., Ltd. Data recording apparatus and method with verification of optical disc
US5359674A (en) * 1991-12-11 1994-10-25 David Sarnoff Research Center, Inc. Pyramid processor integrated circuit
US5448372A (en) * 1991-12-18 1995-09-05 Eastman Kodak Company Storage and retrieval of digitized photographic images
US5521717A (en) * 1992-02-03 1996-05-28 Canon Kabushiki Kaisha Image processing apparatus and method therefor
KR0134177B1 (en) * 1992-03-24 1998-04-22 Toshiba Kk Variable length code recording playback apparatus
GB2267194B (en) * 1992-05-13 1995-10-04 Sony Broadcast & Communication Apparatus and method for processing image data
JP2666662B2 (en) * 1992-06-29 1997-10-22 日本ビクター株式会社 Hierarchical encoding device and decoding device
US6181826B1 (en) 1992-07-22 2001-01-30 Eastman Kodak Company Method and associated apparatus for achieving additional signal level resolution from a quantized digital representation of an image
US5381526A (en) * 1992-09-11 1995-01-10 Eastman Kodak Company Method and apparatus for storing and retrieving generalized image data
EP0610604A1 (en) * 1993-02-11 1994-08-17 Agfa-Gevaert N.V. Multiple processing of radiographic images based on a pyramidal image decomposition
EP0610606A1 (en) * 1993-02-11 1994-08-17 Agfa-Gevaert N.V. Method of displaying part of a radiographic image
EP0610607A1 (en) * 1993-02-11 1994-08-17 Agfa-Gevaert N.V. Method of evaluating image processing performed on a radiographic image
EP0610603B1 (en) * 1993-02-11 1999-09-08 Agfa-Gevaert N.V. Fast interactive off-line processing method for radiographic images
EP0610602B2 (en) * 1993-02-11 2010-11-24 Agfa HealthCare N.V. Method of displaying (a) radiographic image(s)
EP0614308A1 (en) * 1993-03-05 1994-09-07 Eastman Kodak Company Method and apparatus for controlling access to selected image components in an image storage and retrieval system
US5477345A (en) * 1993-12-15 1995-12-19 Xerox Corporation Apparatus for subsampling chrominance
US6084978A (en) * 1993-12-16 2000-07-04 Eastman Kodak Company Hierarchical storage and display of digital images used in constructing three-dimensional image hard copy
US5577134A (en) * 1994-01-07 1996-11-19 Panasonic Technologies, Inc. Method and apparatus for encoding a segmented image without loss of information
EP0701756B1 (en) * 1994-03-18 1999-12-29 Koninklijke Philips Electronics N.V. Audio-visual presentation system
EP0684565A1 (en) * 1994-05-27 1995-11-29 Eastman Kodak Company Medical image archiving of lossy and lossless images on a recordable CD
JPH0865659A (en) * 1994-05-27 1996-03-08 Eastman Kodak Co Method for generating compression data having loss in courseof upload of lossless data
EP0684567A1 (en) * 1994-05-27 1995-11-29 Eastman Kodak Company Dual operation modes for palindrome and lossy/lossless display of medical images
US5724582A (en) * 1994-05-27 1998-03-03 Eastman Kodak Company Medical image archiving with lossy images on two or more recordable CDs
JP3248186B2 (en) * 1994-08-31 2002-01-21 ソニー株式会社 Still image system
JPH08110950A (en) * 1994-09-08 1996-04-30 Sony Corp Plotting device and method utilizing hierarchical approximation system for graphic data
US5568570A (en) 1994-09-30 1996-10-22 Eastman Kodak Company Method and apparatus for reducing quantization artifacts in a hierarchical image storage and retrieval system
US5619590A (en) * 1994-10-06 1997-04-08 Eastman Kodak Company System for electronic image signal processing to provide a tonescale corrected full resolution luminance and two half resolution chrominance signals
US7362775B1 (en) 1996-07-02 2008-04-22 Wistaria Trading, Inc. Exchange mechanisms for digital information packages with bandwidth securitization, multichannel digital watermarks, and key management
US5835637A (en) * 1995-03-20 1998-11-10 Eastman Kodak Company Method and apparatus for sharpening an image by scaling spatial residual components during image reconstruction
EP0741371B1 (en) * 1995-05-01 2002-10-23 Agfa-Gevaert Radiation image reproducing method
JP4174083B2 (en) * 1995-05-01 2008-10-29 キヤノン株式会社 Image processing method and apparatus
US5613004A (en) 1995-06-07 1997-03-18 The Dice Company Steganographic method and device
US5896465A (en) * 1995-06-20 1999-04-20 Harris Corporation Reverse prioritized image transmission
JP3785700B2 (en) 1995-12-18 2006-06-14 ソニー株式会社 Approximation method and apparatus
US6205249B1 (en) 1998-04-02 2001-03-20 Scott A. Moskowitz Multiple transform utilization and applications for secure digital watermarking
US7664263B2 (en) 1998-03-24 2010-02-16 Moskowitz Scott A Method for combining transfer functions with predetermined key creation
US5764235A (en) * 1996-03-25 1998-06-09 Insight Development Corporation Computer implemented method and system for transmitting graphical images from server to client at user selectable resolution
KR100464233B1 (en) * 1996-05-08 2005-04-13 코닌클리케 필립스 일렉트로닉스 엔.브이. Apparatus and method for transmitting digital information signal having specific first sampling frequency
US7177429B2 (en) 2000-12-07 2007-02-13 Blue Spike, Inc. System and methods for permitting open access to data objects and for securing data within the data objects
US7095874B2 (en) 1996-07-02 2006-08-22 Wistaria Trading, Inc. Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data
US7123718B1 (en) 1999-03-24 2006-10-17 Blue Spike, Inc. Utilizing data reduction in stegnographic and cryptographic systems
US7457962B2 (en) 1996-07-02 2008-11-25 Wistaria Trading, Inc Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data
US7346472B1 (en) 2000-09-07 2008-03-18 Blue Spike, Inc. Method and device for monitoring and analyzing signals
US7159116B2 (en) 1999-12-07 2007-01-02 Blue Spike, Inc. Systems, methods and devices for trusted transactions
US5889868A (en) 1996-07-02 1999-03-30 The Dice Company Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data
JPH10134208A (en) 1996-10-31 1998-05-22 Sony Corp Shape data approximating method and plotting device
JP3785709B2 (en) * 1996-12-13 2006-06-14 ソニー株式会社 Shape data approximation method and drawing apparatus
US7730317B2 (en) 1996-12-20 2010-06-01 Wistaria Trading, Inc. Linear predictive coding implementation of digital watermarks
JPH10233995A (en) * 1997-02-20 1998-09-02 Eastman Kodak Japan Kk Electronic still camera and its reproduction display method
US6021224A (en) * 1997-03-28 2000-02-01 International Business Machines Corporation Multiresolution lossless/lossy compression and storage of data for efficient processing thereof
JPH114367A (en) * 1997-04-16 1999-01-06 Seiko Epson Corp High speed image selection method and digital camera with high speed image selection function
JPH10304334A (en) * 1997-04-25 1998-11-13 Canon Inc Communication method and device, transmission device and reception device, communication system and recording medium
JP2000067270A (en) 1998-06-12 2000-03-03 Sony Corp Method for approximating shape data, information processor and medium
US6208770B1 (en) 1998-09-18 2001-03-27 Eastman Kodak Company Digital colored corrected prints produced from colored film
EP1172001A4 (en) * 1999-03-24 2002-08-14 Blue Spike Inc Utilizing data reduction in steganographic and cryptographic systems
US7664264B2 (en) 1999-03-24 2010-02-16 Blue Spike, Inc. Utilizing data reduction in steganographic and cryptographic systems
US6614471B1 (en) 1999-05-10 2003-09-02 Banctec, Inc. Luminance correction for color scanning using a measured and derived luminance value
US7221782B1 (en) 1999-06-24 2007-05-22 General Electric Company Method and apparatus for determining a dynamic range of a digital medical image
US6704440B1 (en) 1999-06-24 2004-03-09 General Electric Company Method and apparatus for processing a medical image containing clinical and non-clinical regions
US6460003B1 (en) 1999-07-01 2002-10-01 General Electric Company Apparatus and method for resolution calibration of radiographic images
US6633657B1 (en) 1999-07-15 2003-10-14 General Electric Company Method and apparatus for controlling a dynamic range of a digital diagnostic image
WO2001018628A2 (en) 1999-08-04 2001-03-15 Blue Spike, Inc. A secure personal content server
US6785423B1 (en) * 2000-05-26 2004-08-31 Eastman Kodak Company Producing a compressed digital image organized into layers having information relating to different viewing conditions and resolutions
US7127615B2 (en) 2000-09-20 2006-10-24 Blue Spike, Inc. Security based on subliminal and supraliminal channels for data objects
JP3969981B2 (en) * 2000-09-22 2007-09-05 キヤノン株式会社 Electron source driving method, driving circuit, electron source, and image forming apparatus
JP3969985B2 (en) 2000-10-04 2007-09-05 キヤノン株式会社 Electron source, image forming apparatus driving method, and image forming apparatus
US7133597B2 (en) * 2001-07-05 2006-11-07 Eastman Kodak Company Recording audio enabling software and images on a removable storage medium
US7287275B2 (en) 2002-04-17 2007-10-23 Moskowitz Scott A Methods, systems and devices for packet watermarking and efficient provisioning of bandwidth
KR100518849B1 (en) * 2004-03-02 2005-09-30 엘지전자 주식회사 Image coding and decoding method
GB0514555D0 (en) 2005-07-15 2005-08-24 Nonlinear Dynamics Ltd A method of analysing separation patterns
GB0514553D0 (en) * 2005-07-15 2005-08-24 Nonlinear Dynamics Ltd A method of analysing a representation of a separation pattern
US20080212883A1 (en) * 2005-08-17 2008-09-04 Pixoneer Geomatics, Inc. Processing Method of Data Structure for Real-Time Image Processing
US7571064B2 (en) * 2006-03-31 2009-08-04 Apple Inc. Display digital signal visualizations with increasing accuracy
FR2909474B1 (en) * 2006-12-04 2009-05-15 Canon Kk METHOD AND DEVICE FOR ENCODING DIGITAL IMAGES AND METHOD AND DEVICE FOR DECODING CODE DIGITAL IMAGES
GB2445008B (en) * 2006-12-20 2008-12-31 Sony Comp Entertainment Europe Image compression and/or decompression
US8063901B2 (en) * 2007-06-19 2011-11-22 Siemens Aktiengesellschaft Method and apparatus for efficient client-server visualization of multi-dimensional data
WO2009032255A2 (en) * 2007-09-04 2009-03-12 The Regents Of The University Of California Hierarchical motion vector processing method, software and devices
US8391638B2 (en) * 2008-06-04 2013-03-05 Microsoft Corporation Hybrid image format
GB2491448B (en) 2011-05-05 2014-05-28 Advanced Risc Mach Ltd Apparatus for and method of dividing data for encoding and decoding purposes
US9886945B1 (en) * 2011-07-03 2018-02-06 Reality Analytics, Inc. System and method for taxonomically distinguishing sample data captured from biota sources
JP5474887B2 (en) 2011-08-01 2014-04-16 株式会社ソニー・コンピュータエンタテインメント Moving image data generation device, moving image display device, moving image data generation method, moving image display method, and data structure of moving image file
US10147202B2 (en) * 2013-03-15 2018-12-04 Arm Limited Methods of and apparatus for encoding and decoding data
US9489104B2 (en) 2013-11-14 2016-11-08 Apple Inc. Viewable frame identification
US9582160B2 (en) 2013-11-14 2017-02-28 Apple Inc. Semi-automatic organic layout for media streams
US20150254806A1 (en) * 2014-03-07 2015-09-10 Apple Inc. Efficient Progressive Loading Of Media Items
US10303498B2 (en) 2015-10-01 2019-05-28 Microsoft Technology Licensing, Llc Performance optimizations for emulators
US11403100B2 (en) 2020-08-31 2022-08-02 Microsoft Technology Licensing, Llc Dual architecture function pointers having consistent reference addresses
US11042422B1 (en) 2020-08-31 2021-06-22 Microsoft Technology Licensing, Llc Hybrid binaries supporting code stream folding
US11231918B1 (en) 2020-08-31 2022-01-25 Microsoft Technologly Licensing, LLC Native emulation compatible application binary interface for supporting emulation of foreign code

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0152355A2 (en) * 1984-02-14 1985-08-21 Nippon Telegraph And Telephone Corporation Image information retrieval/display apparatus
US4829453A (en) * 1987-03-05 1989-05-09 Sharp Kabushiki Kaisha Apparatus for cataloging and retrieving image data

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4674125A (en) * 1983-06-27 1987-06-16 Rca Corporation Real-time hierarchal pyramid signal processing apparatus
JPS60148279A (en) * 1983-12-28 1985-08-05 インタ−ナショナル　ビジネス　マシ−ンズ　コ−ポレ−ション Image processing system
GB8429879D0 (en) * 1984-11-27 1985-01-03 Rca Corp Signal processing apparatus
US4709394A (en) * 1985-08-23 1987-11-24 Rca Corporation Multiplexed real-time pyramid signal processing system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0152355A2 (en) * 1984-02-14 1985-08-21 Nippon Telegraph And Telephone Corporation Image information retrieval/display apparatus
US4829453A (en) * 1987-03-05 1989-05-09 Sharp Kabushiki Kaisha Apparatus for cataloging and retrieving image data

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
FUNK-TECHNIK. vol. 41, no. 12, December 86, HEIDELBERG DE pages 501 - 503; C.Reuber: "Digitale optische Speicherplatten der dritten Generation" see abstract *
IBM TECHNICAL DISCLOSURE BULLETIN. vol. 31, no. 9, February 1989, NEW YORK US pages 30 - 34; "Memory architecture supporting multiple image formats" see page 1; figure 1 *
IEEE TRANSACTIONS ON COMMUNICATIONS. vol. 31, no. 4, April 1983, NEW YORK US pages 532 - 540; P.J.Burt and E.H.Adelson: "The Laplacian Pyramid as a compact image code" see abstract; figures 4, 5 (cited in the application) *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6370274B1 (en) 1993-10-14 2002-04-09 U.S. Philips Corporation Method of storing in coded form image information of moving images, diagnostic system using the method, and image coding and recording unit and image-information retrieval unit for use in such a system

Also Published As

Publication number Publication date
JPH04503295A (en) 1992-06-11
DE69029725D1 (en) 1997-02-27
EP0455794A1 (en) 1991-11-13
EP0455794B1 (en) 1997-01-15
US4969204A (en) 1990-11-06
DE69029725T2 (en) 1997-07-03

Similar Documents

Publication Publication Date Title
US4969204A (en) Hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment
US5050230A (en) Hybrid residual-based hierarchical storage and display method for high resolution digital images in a multiuse environment
US5048111A (en) Hybrid subband-based hierarchical storage and display method for high resolution digital images in a multiuse environment
EP0516826B1 (en) Metric conversion mechanism for digital images in a hierarchical, multi-resolution, multi-use environment
EP0460191B1 (en) Methods for reducing quantization error in hierarchical decomposition and reconstruction schemes
US8054886B2 (en) Signaling and use of chroma sample positioning information
US4663660A (en) Compressed quantized image-data transmission technique suitable for use in teleconferencing
JP2677574B2 (en) Data display device
US6665343B1 (en) Methods and arrangements for a converting a high definition image to a lower definition image using wavelet transforms
JPS62250770A (en) Method of encoding image and method of demodulation and apparatus for realizing the methods
CN1246233A (en) A multiple format video signal processor
US6819801B2 (en) System and method for processing demosaiced images to reduce color aliasing artifacts
US4858026A (en) Image display
Segall Scalable coding of high dynamic range video
KR20220053657A (en) Tensor-product B-spline predictors
JP2529709B2 (en) Image display method and device
US6181826B1 (en) Method and associated apparatus for achieving additional signal level resolution from a quantized digital representation of an image
JPH05191653A (en) Method for coding and decoding color image and coder and decoder using this method
US6445870B1 (en) Image compression and expansion device
JP3260284B2 (en) Image compression device and image decompression device
JPH1141597A (en) Binary form signal restoring method
JPH0678297A (en) Method for encoding of digital video signal
JPH07322255A (en) Hierarchical coder for digital image signal
JP2937057B2 (en) Digital composite video signal encoding device and decoding device
JPH03182171A (en) Coding and reproducing method or color picture information

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FR GB GR IT LU NL SE

WWE Wipo information: entry into national phase

Ref document number: 1991900444

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1991900444

Country of ref document: EP

WWG Wipo information: grant in national office

Ref document number: 1991900444

Country of ref document: EP