WO2016180407A1

WO2016180407A1 - Method for determining an individual sharpness of an image of an iris, and method for people screening

Info

Publication number: WO2016180407A1
Application number: PCT/DE2016/100218
Authority: WO
Inventors: Ralf Grieser; Andreas Wolf
Original assignee: Bundesdruckerei Gmbh
Priority date: 2015-05-12
Filing date: 2016-05-11
Publication date: 2016-11-17
Also published as: EP3295369A1; DE102015107474B4; DE102015107474A1

Abstract

The disclosure describes a method for determining an individual sharpness of an image of an iris, comprising: recording an image of an eye; and storing the image in an image memory; comprising in the image memory: finding the pupil of an eye; determining at least two sampling regions which are each arranged partly within the pupil and partly outside of the pupil, with each sampling region, on its own, being connected; establishing a reference sampling region which has a maximum reference contrast that is greater than the contrast of each other sampling region; storing the position of the reference sampling region and the maximum reference contrast; recording a further image of the eye under variation of the sharpness setting; storing the further image in the image memory; establishing the current contrast in the reference sampling region; comparing whether the current contrast is greater than a preceding contrast; and carrying out the steps until the current contrast is the greatest achievable one. Moreover, the application relates to a method for people screening and a method for determining a digitized image recording of an iris in a data processing device.

Description

Method for determining an individual sharpness of an image of an iris

and method for personal examination

The invention relates to technologies in the context of determining an individual sharpness of an image of an iris, as well as the use.

background

It is known to make a determination of the person's iris in the context of biometry, in particular for person identification. Iris determination can also be used in animals. Digitized image recordings of the iris are evaluated by means of electronic data processing in order to carry out the determination. For digital image analysis, the aim is to have as good as possible processable image recordings, wherein the image quality is influenced in particular by the sharpness of the image recording, which depends on the focus setting of the image recording device. The sharpness of the image is a measure of the distinctiveness of details in the image capture. There is therefore an attempt to obtain the sharpest possible image recordings of the eye for the iris determination, which required a necessary focus adjustment of the image recording device. Document EP 1 171 996 B1 discloses a method for determining focus in an image recording device. On the digitized images, which in particular can also show an iris image, a convolution with a special convolution kernel is applied to the pixels, which corresponds to the application of a high-pass filter. In this way, the proportion of high frequencies of the image recording is to be emphasized compared to lower frequencies, since the high frequencies are particularly suitable for deriving a measure of the sharpness of the recorded images and thus the focus setting of the image recording device.

There are cameras with automatic focusing with contrast comparison known (Hussman: How does a contrast comparison autofocus?; In: http://digicam-experts.de; 12.04.2015, 1-3.) Http://www.archive.org/web/ 20150412203640 / http: //digicam-experts.de/wissen/25). The document DE 26 58 832 AI relates to motion picture cameras with automatic focusing device whose drive is stopped or interrupted by the contrast measuring device for the image designed by the lens when reaching the Konstrastmaximums.

Document DE 1 279 374 A relates to a method and a device for automatically adjusting the shades of optical systems by contrast measurement and conversion of the measured values into electrical energy. Document US 5,659,812 relates to determining an optimized focus area by means of an algorithm.

Document DE 44 90 836 Tl relates to personal identification instruments and, more particularly, to a method of making such an instrument which is highly tamper-proof.

The document DE 10 2007 055 923 A1 relates to a method for determining properties or determining and / or tracking properties of the position of the center of characteristic eye components, in particular of the limbus, during an eye examination or treatment. In this case, a multicolored digital image of at least one section of an eye is taken with the camera and based on this the position or the radius is derived using image processing methods. According to the invention, the red channel of the digital image is used as the basis for the image processing. Summary

The object of the invention is to provide a method for determining a digitized image recording of an iris in a data processing device and uses thereof, in which iris images in a simple and efficient manner with the least possible expenditure of time can be evaluated. According to another aspect, the personal examination should be made more efficient.

As a solution, a method for determining an individual sharpness of an image of an iris according to independent claim 1 is provided. Furthermore, a method for personal examination according to claim is created. Furthermore, a method for determining a digitized image recording of an iris in a data processing device according to the subordinate claim 3 is provided. Claims 9 and 10 relate to uses of the method. Embodiments are the subject of dependent claims.

In one aspect, a method is provided for determining an individual sharpness of an image of an iris. The method comprises the following steps:

a) taking a picture of an eye; and

b) storing the image in a frame buffer;

in image memory having:

c) finding a pupil of the eye;

d) determining at least two scan areas, each partially located in the pupil and partially external to the pupil, each scan area being coherent in its entirety;

e) determining a reference sample area having a maximum reference contrast which is greater than the contrast of each other sample area;

f) storing the location of the reference scan area and the maximum reference contrast; g) taking a further image of the eye while varying a focus;

h) storing the further image in the image memory;

i) determining the current contrast in the reference scan area;

j) comparing whether the current contrast is greater than a previous contrast; and k) performing steps g) to i) until the current contrast is the highest achievable. In another aspect, a method of personal verification is proposed. The procedure includes the following steps:

A) capturing an individual image of an individual iris, in particular by means of a method for determining an individual sharpness of an image, as well as at least one further biometric individual feature of a person and forming a combined biometric data set from the individual image and the further biometric individual feature;

B) creating a database with a plurality of combined biometric datasets;

C) classifying the combined biometric data sets in the database so as to obtain classified combined biometric data sets each consisting of a classified image and a classified further biometric feature; and

D) when acquiring a measuring image of an iris of a person by means of the method for determining an individual sharpness of an image of an iris:

E) comparing the measurement image with the images of the classified combined biometric datasets and, based thereon, finding a best-match biometric dataset;

F) determining the classified further biometric characteristic of the Best Match biometrics dataset;

G) comparing the further biometric feature of the person with the classified further biometric feature of the best match biometric data set; and

H) issuing a warning if the person's further biometric feature deviates from the classified further biometric feature of the best-match biometric dataset by a predetermined amount.

In the method for determining an individual sharpness of an image of an iris and / or in the method for person verification, alternatively or additionally to the contrast, a contrast increase (image contrast increase) can be determined for the scanning regions and used as a measure of the image sharpness.

In another aspect, there is provided a method of determining a digitized image capture of an iris in a data processing device. The data processing device is set up by means of a software application for digital image processing. In the driving an iris image is provided in digitized form. The iris image is an image of an eye showing at least the area of the iris and the pupil of the eye. An image section or section of the iris image (scanning region) is selected, which shows a transition section with a pupil region and an iris region adjacent thereto. An image contrast increase is determined for a transition between the pupil area and the adjacent iris area in the transition section. From the increase in the image contrast, or in short contrast increase, a sharpness measure for the iris image is determined.

From an analysis of the contrast properties of the iris image in the area of the transition between a pupil area and an adjacent iris area, a sharpening dimension of the iris image is thus determined according to the various aspects, which indicates a measure of the focus setting of the image recording device used for image acquisition.

The image contrast increase can be a local contrast increase, which, when applying contrast values above a local dimension for the pixels in the minimum case, characterizes the increase in contrast over a few pixels, for example at least three pixels. By way of example, the values which represent a measure of the brightness of the pixels can be plotted as y values, wherein x values each represent a pixel or channel of the image recording device, for example equidistantly spaced points. When comparing the thus determined gradient / increase of the contrast for different images of the same transition section or scanning region, the curves to be compared can be steeper or flatter, indicating different contrast increases in the examined region.

A local group of pixels can also be included in determining the local contrast increase or the local contrast, one or more pixels being assigned to the pupil region and one or more pixels to the adjacent iris region. The pixels can lie on a straight line, which intersects the border area between the pupil area and the adjacent iris area. In one embodiment it can be provided that the sharpness dimension is determined as being proportional to the ascertained image contrast increase between the pupil area and the adjacent iris area.

The method can be used in connection with a biometric determination of an iris or in an iris image recording device.

When determining the sharpness dimension for digital pixels of the transition, FFT values (FFT - Fast Fourier Transformation) can be determined or determined between the pupil region and the adjacent iris region by means of fast Fourier transformation.

When determining the sharpness, the FFT values of an algebraic operation can be summarized accordingly. The FFT values along a transition between the pupil area and the iris area may be added in one embodiment. The contrast increase in the transition region may correspond to the difference of the FFT values of the pixels on both sides of the transition between the pupil region and the iris region. Additionally or alternatively, the slope of the curve of the FFT values in the transition region can be determined. The measure of the sharpness of the iris image may be proportional to the difference and / or increase. Proportionality with respect to the sum of the FFT values may also be provided for the sharpness measure.

When determining the image contrast increase, an edge can be determined in the course of the pixel values assigned to the pixels.

It can also be provided that

a further image section of the iris image is selected which shows, along the circumference of the pupil, another transition section with a pupil region and an iris region bordering on it,

a further image contrast increase is determined for the further transition between the pupil region and the adjacent iris region in the transition section, and - the image contrast increase and / or further image contrast increase are selected to determine therefrom a respective degree of sharpness for the iris image.

In this way, the contrast increase can be determined for different image sections which are arranged along the circumference of the pupil. In this case, the determination of the image contrast increase can be provided for adjacent and / or spaced transition sections along the circumference of the pupil. On the basis of the respectively determined image contrast increase, a sharpening measure can then be derived. In this way, it may be provided in one embodiment to determine a transition section along the circumference of the pupil, which will then be used for further analyzes, in particular the determination of a sharpening measure for the iris image. In an embodiment, it can be provided that, starting from the determination of the transitional section between a pupil area and an adjacent iris area, door which has been determined the greatest increase in contrast, a more extensive image analysis is carried out.

The contrast determination for several transition sections along the circumference of the pupil can be performed for one or more iris images, all of which show the same iris. Thereafter, it may be provided to select one of the transition sections for the sharpness measurement. The focus can in turn be determined for one and the same transition section along the circumference of the pupil for several iris images, so that ultimately an iris image with the best or highest sharpening measure can be determined. This iris image can then be supplied, for example, to further image processing or processing, for example for iris determination in the context of biometry. In one embodiment, it may be provided that

a further iris image is provided in digitized form, which shows the same iris as the iris image,

an image section of the further iris image is selected, which shows the same transition section with the pupil region and the adjacent iris region along the circumference of the pupil, an image contrast increase for the transition between the pupil area and the adjacent iris area in the transition section of the further iris image is determined, and

- A sharpness measure for the other iris image is determined from the image contrast increase.

For each of the several iris images, a specific measure of the sharpness of the image can be compared in order to select an iris image suitable for further image processing or processing from the set of iris images examined.

The illustrated embodiments may be provided individually or in any combination in conjunction with the alternative method embodiments.

Description of exemplary embodiments

In the following, further exemplary embodiments are explained with reference to figures of a drawing. Hereby show:

1 shows a schematic representation of an arrangement for recording and evaluating digital image recordings showing at least the area with the iris and the pupil from an eye to be determined,

2 is a schematic representation of an iris with pupil,

3 shows a flowchart for a method for determining iris images in digital form, FIG. 4 shows a graph of a standardized intensity for pixels along an image-capturing line,

5 shows a graphical representation of the image intensity over a series of pixels in a transition section between a pupil region and an iris region for iris images with different image sharpness,

6 is a graph of FFT values for the waveforms of FIG. 5, and FIG. 7 is a graph of accumulated FFT values for iris images of differing sharpness. Fig. 1 shows a schematic representation of an arrangement with an image pickup device 1, which is designed for example as a digital camera. The image recording device I is arranged opposite an eye 2 from which digital image recordings are to be made with the aid of the image recording device 1, which images show at least the region of the iris with the pupil of the eye 2. The image recording device I is functionally coupled to a data processing device 3, which is set up with the aid of a software application installed thereon for processing and processing the digital image recorded by the image recording device 1. In this case, provision may be made for control signals to be generated in the data processing device 3 as a function of a performed image analysis, which serve to control the image recording device 2, for example for controlling a triggering time for the image recording and / or a positioning of the image recording device 1 Reference to the eye 2, which is shown schematically in Fig. I with the aid of an arrow A. Alternatively or additionally, it can be provided that audible and / or visual information is output to the user so that the latter displaces his eye 2 in relation to the image recording device 1. This too can be controlled as a function of image processing in the data processing device 3.

Alternatively or additionally, optical components, for example lenses and / or a lens, can be set in the image recording device 2 as a function of control signals for image acquisition for focus adjustment for the image recordings, wherein the control signals can be the result of an image analysis in the data processing device 3 , Thus, a measure of the sharpness of the recorded images can be determined by means of the data processing device 3, in order then to give a focus control signal to the image recording device 2 from the comparison of the sharpness dimension for several images and / or to select one of the examined images for a further image evaluation, for example, for a biometric analysis for person recognition or verification. Data processing device 3 and image recording device 2 can also be integrated in a common housing. The arrangement shown can be used, for example, for person identification in the context of a biomedical method.

FIG. 2 shows a schematic representation of a section of an iris image with pupil 20 and 5 iris 21. In the image analysis, one or more local transitions 22 between pupil 20 and iris 21 can be evaluated in order to obtain a respective local increase in contrast between the image area of the pupil 20 and to determine the adjacent image area for the iris 21. A circumferential boundary region 23 between pupil 20 and iris 21 can be partially or completely detected in the image evaluation by displacing the local transition 22 along the boundary region 230.

With reference to FIG. 3, a method for determining one or more iris images in digitized form will be explained below. In step 30, an iris image is taken with the aid of the image recording device 2 and provided for image analysis in the data processing device 3. With the aid of the software application implemented on the data processing device 3 for the image analysis, an image section (scanning region) is selected in step 31, which comprises a transition section between a pupil region 20 and an iris region 21 (see FIG. FIG. 4 shows by way of example a standardized measured intensity for pixels of the iris image along a line B in the image representation from FIG. 4 (see also FIG. 2). The intensity decreases in a central region 40, which corresponds to the pupil 20. In the region 41, the intensity of the transition to the area of the iris 21 increases steeply. Fig. 5 shows this transition in detail. Curve 50 corresponds to the course of the measured curve in FIG. 4. Furthermore, a curve 51 is shown in which the steepness of the curve in the rise is significantly lower, which corresponds to a different contrast increase. The curve 51 shows the course for the same local transition between the pupil 20 and the iris 21 in a so-called Blurr image, which was derived for comparison purposes from the measured values (see FIG. The a lower image sharpness in the area between the pupil area 20 and the iris area 21 leads to a changed curve increase, the actual contrast (brightness difference) being the same here.

In step 32, a fast Fourier transform is then applied to the digital images to determine FFT values for the waveforms shown in FIG. The result of this operation is shown in FIG. 6 where the curves 60 and 61 show the FFT values for the curves 50 and 51.

In the image analysis in the data processing device 3, a measure of the sharpness of the analyzed iris images is determined based on the determined increase in contrast (step 33). For example, in this case the steepness in the transition region between pupil 20 and iris region 21 can be used.

Alternatively or additionally, the FFT values for the curve profiles in FIG. 6 can be summed up. The result of this operation is shown in FIG. 7 in the form of the values 70 and 71 for the two curves 60 and 61. In FIG. 7, two further results 72,..., 74 are entered. For these, the Blurr radius was widened, simulating a further decrease in image sharpness.

The method described above can be used to determine the best focus image pick-up in a sequence of images taken for one and the same iris, which is the best focus adjustment of the image pickup device 1 in relation to the image pickup device Eye 2 corresponds. The iris image selected in this case can then be used for further image analysis, for example for iris biometry, which can also be carried out in the data processing device 3, for example.

The described method can also be used with all the steps or parts thereof in order to control the process of image recording with the aid of the image recording device 2, for example in such a way that in an initial measuring phase a plurality of iris images for different distances between the eye 2 and the image recording device 1 be taken to in this way a preferred distance between the measuring device and the To determine measurement object (eye), in order to make the actual iris image recording. The iris image recorded in this case can then be analyzed, for example for the purpose of an iris biometry. Thus, an improved method for person identification by means of iris biometry is provided in various ways.

The features disclosed in the above description, the claims and the drawings may be of significance both individually and in any combination for the realization of the various embodiments.

Claims

Bundesdruckerei G

B60614WO

claims

A method of determining an individual sharpness of an image of an iris, comprising: a) taking an image of an eye; and

b) storing the image in a frame buffer;

in image memory having:

c) finding a pupil of the eye;

d) determining at least two scan areas, each located partially in the pupil and partially outside the pupil, each scan area being contiguous in its own right;

f) storing the position of the reference scanning range and the maximum reference contrast;

g) taking a further image of the eye while varying a focus;

h) storing the further image in the image memory;

i) determining the current contrast in the reference scan area;

j) comparing whether the current contrast is greater than a previous contrast; and k) performing steps g) to i) until the current contrast is the greatest achievable.

2. A method of checking persons, comprising:

A) acquiring an individual image of an individual iris with the method according to claim 1 and at least one further biometric individual feature of a person and forming a combined biometric data set from the individual image and the further biometric individual feature;

B) creating a database with a plurality of combined biometric datasets;

C) classifying the combined biometric data sets in the database so as to obtain classified combined biometric data sets each consisting of a classified image and a classified further biometric feature; and D) when acquiring a measurement image of an iris of a person with the method according to claim I:

E) comparing the measurement image with the images of the classified combined biomimetic datasets and, based thereon, finding a best-match biometric dataset;

F) determining the classified further biometric feature of the best match biometric dataset;

A method for determining a digitized image capture of an iris in a data processing device that is set up by means of a software application for digital image processing, the method comprising the following steps:

Providing an iris image in digitized form,

Selecting an image section of the iris image which shows a transition section with a pupil region and an iris region adjacent thereto,

Determining an image contrast increase for a transition between the pupil area and the adjacent iris area in the transition section and

- Determining a sharpness measure for the iris image from the image contrast increase.

Method according to Claim 3, characterized in that FFT values are determined by means of fast Fourier transformation when determining the sharpness dimension for digital pixels of the transition between the pupil region and the adjacent iris region. A method according to claim 4, characterized in that the FFT values of an algebraic operation are summarized when determining the sharpness measure.

Method according to at least one of claims 3 to 5, characterized in that, when determining the image contrast increase, an edge is determined in the course of pixel values associated with the pixels.

Method according to at least one of claims 3 to 6, characterized in that

a further image section of the iris image is selected which shows, along the circumference of the pupil, another transition section with a pupil region and a neighboring iris region,

a further image contrast increase is determined for the further transition between the pupil area and the adjacent iris area in the transition section, and

the image contrast increase and / or further image contrast increase are selected in order to determine therefrom a respective focal length for the iris image.

Method according to at least one of claims 3 to 7, characterized in that

an image section of the further iris image is selected, which shows the same transition section with the pupil region and the adjacent iris region along the circumference of the pupil,

an image contrast increase for the transition between the pupil area and the adjacent iris area in the transition section of the further iris image is determined, and

9. Use of a method according to at least one of claims 3 to 8 in a biometric determination of an iris. 10. Use of a method according to any one of claims 3 to 8 in an iris image pickup device.