KR100551775B1 - Discrimination method of imitative iris in iris recognition system - Google Patents

Abstract

The present invention relates to a method for discriminating a fake iris in an iris recognition system.

The imitation iris discrimination method of the present invention distinguishes a real human eye from a fake eye based on the geometrical characteristics of the pupil, even when the brightness of the eye image input from the camera is adjusted. To this end, the present invention uses the brightness or contrast of the camera itself, the shutter speed, or the image obtained by adjusting the brightness of the eye image by using the brightness or contrast in the image acquisition board (image capture board), or the brightness change of the lighting device itself, etc. The imitation iris is accurately identified using the difference in the geometrical characteristics of the real eye and the fake eye appearing in the difference image of the two images in which the brightness of the pupil part is changed.

Iris recognition, imitation iris, discrimination

Description

Counterfeit iris discrimination method in iris recognition system {DISCRIMINATION METHOD OF IMITATIVE IRIS IN IRIS RECOGNITION SYSTEM}

1 is a view showing a schematic configuration of an iris recognition system

2 is a flowchart showing an operation procedure of the iris recognition system.

3 is a diagram for explaining a method of discriminating a fake iris using Fourier spectrum analysis of an iris image;

4 is a flowchart of an embodiment of a method for discriminating a counterfeit iris according to the present invention;

5 is a view showing a schematic structure of an iris recognizer according to the dummy iris discrimination method of the present invention

Figure 6 is a view for explaining a pupil change in accordance with the brightness change of the wearer in the present invention

7 is a view for explaining a pupil change in accordance with the change in brightness in the case of a dummy eye in the present invention

8 is a view for explaining the pupil change according to the brightness change of the wearer in the present invention;

The present invention relates to a method for determining the authenticity of irises (images) photographed by an image acquisition device such as a camera in an iris recognition system. Particularly, the present invention relates to iris recognition by determining the authenticity of irises based on the geometric characteristics of the pupil. The present invention relates to a counterfeit iris discrimination method capable of real-time processing of authenticity discrimination and to achieve more accurate counterfeit discrimination.

As is known, it is a system for security, crime prevention, and identity authentication. It goes beyond the existing contact or contactless card system to recognize human fingerprints, or to authenticate identity through iris recognition and access to specific places or access to specific information. Systems to allow or deny such have been prevalent.

Among them, the iris recognition system has the advantage of higher recognition rate and more accurate recognition than the fingerprint recognition, but when using the imitation iris image using the print or photograph of the registered iris pattern, it is not the actual human iris. In addition, there is a risk of error that is recognized as registered iris.

1 is a view showing the configuration of a general iris recognition system, a camera 2 for capturing an eye image when the human eye 1 is within a predetermined operating region, and an eye photographed by the camera 2. Iris image processing unit 3 for extracting the iris image by processing the image signal and authenticity, and iris image processed by the iris image processing unit 3, using the registered iris database to recognize the iris An iris recognition unit 4 and a distance measuring unit 5 for measuring the distance between the eye 1 and the camera 2 to determine the approach to the operating area.

Referring to the operation of the iris recognition system of Figure 1 as follows.

The distance measuring unit 5 measures the distance between the user's eye 1 and the camera 2 so as to determine whether the user has entered a predetermined operating area, and then zooms or focuses the camera 2 at that time. The function is driven to photograph the user's eye 1.

The eye image photographed by the camera 2 is input to the iris image processor 3, and the iris image processor 3 processes the eye image to extract an iris image. In addition, the authenticity of the extracted iris image is determined, and in the case of a simulated iris, the fact is notified through a display device or a speaker, and in the case of an actual human iris, the result is notified to the iris recognition unit 4. The iris recognition unit 4 receives the result determined by the iris image processing unit 3 as an actual human iris, performs an iris recognition process using a database, checks whether the user is registered, and registers the registered user. In case of access, it is allowed to enter, otherwise it is denied.

As a method for screening the authenticity of the iris, that is, the imitation iris (image), conventionally, a method of examining the degree of contraction and swelling of the pupil or a two-dimensional Fourier Spectrum analysis of the iris image are used. A method is proposed.

In the former case, the pupil uses a feature of repeating contraction and expansion in a constant light or an environment in which the illumination changes, and FIG. 2 shows a procedure of a general iris recognition and a fake eye discrimination method.

In the first step S1, the eye image photographed by the camera is input, and in the next step S2, the pupil and the iris region are extracted from the eye image. And it is determined whether the recognition is performed by performing the recognition process (S3) of the pupil and the iris (S4). If the pupil and the iris are recognized, it is an eye image. At this time, a fake eye discrimination algorithm (using the contraction and dilation features of the pupil) is performed and the result is determined (steps S5 and S6). If the result of the determination is imitation eyes, the process (S1 to S6) is repeated two to three times. If the above procedure is repeated a predetermined number of times, if it is determined to be a fake eye or if iris recognition is not performed, a message (discrimination result) indicating that it has not been recognized is output.

The method starts on the assumption that there is no change in the contraction and expansion of the pupil in the case of the iris image printed or printed. However, in order to detect the contraction and expansion of the pupil, a series of operations to find the position of the pupil and to obtain the radius of the pupil must be repeated several times. Moreover, to examine the contraction and expansion of the pupil, a dynamic image change detection technique, It is disadvantageous to obtain an iris image photographed at least three consecutively and go through the analysis of the iris images. This not only makes the identification of the imitation iris a long time, making it difficult to process in real time, but also some errors in the pupil and iris extraction performance itself, which do not accurately pinpoint the change in the size of the pupil. It remained.

On the other hand, the latter method using the two-dimensional Fourier Spectrum analysis of the iris image is a method using the detection of the characteristic frequency components of the printed iris image, generally when the iris image is output using a printer In consideration of the fact that there is a certain amount of print noise components, the method extracts the printed noise components by frequency analysis and selects the dummy iris.

3 shows a two-dimensional Fourier spectrum analysis method. In the case of a printed image of an eye image (iris image) as shown in FIG. 3 (a), four spectrum high frequency characteristics can be determined by the printed noise component as shown in FIG. 3 (b).

That is, since the printed noise components have directionality and periodicity according to the characteristics of the printer, if they are analyzed in the frequency domain through Fourier spectrum analysis, regular high frequency components appear, and when such high frequency components appear, the corresponding iris image is simulated. It is.

However, this method also takes a lot of time to analyze the Fourier Spectrum, making it difficult to process in real time, and even in the case of a real eye image including many eyebrows inside the iris, a high frequency component similar to that of the printed imitation iris image is detected. There is a disadvantage of high concern.

In addition, there are methods for determining the imitation iris, such as measuring a change in the position of light reflection on the cornea, and measuring infrared reactivity. However, there are many limitations in determining the real human eye and fake eye. Concerns about burden and misperception are very high.

SUMMARY OF THE INVENTION An object of the present invention is to provide a counterfeit iris discrimination method capable of accurately screening a counterfeit iris in real time in an iris recognition system.                         

In particular, an object of the present invention is to provide a counterfeit iris discrimination method capable of distinguishing a real human eye from a fake eye by using geometrical characteristics of the pupil in the eye image.

Another object of the present invention is to accurately screen the imitation iris in real time based on the geometric characteristics of the real human eye that even when adjusting the brightness of the eye image input from the camera, the degree of change in the brightness of the pupil is not large. To provide a method of discriminating imitation iris.

In the iris recognition system of the present invention, a method of discriminating a fake iris comprises: acquiring a pupil and an iris image from a photographed eye image, measuring a degree of change in brightness of a pupil part of the acquired pupil and iris image, and the measured pupil Characterized in that it comprises the step of determining whether the imitation iris on the basis of the degree of change in brightness of the portion.

In the imitation iris discrimination method of the iris recognition system of the present invention, it is determined that the pupil part is less than or equal to the threshold set by the brightness gradient, and the case where the brightness gradient is more than or equal to the threshold threshold. Characterized in that it is determined to be an iris.

In the imitation iris discrimination method of the iris recognition system of the present invention, two eye images photographed at different times by two lighting means are used to obtain a degree of change in brightness of the pupil part.

In the imitation iris discrimination method of the iris recognition system of the present invention, a difference image of two eye images taken at two different time intervals is used to obtain a degree of brightness variation of the pupil part.

In addition, in the imitation iris discrimination method of the iris recognition system of the present invention, obtaining eye images having different brightnesses by adjusting brightness, contrast, shutter speed, etc. in the camera itself to obtain the degree of change in brightness of the pupil part. It features.

In addition, in the imitation iris discrimination method of the iris recognition system of the present invention, in order to obtain the degree of change in brightness of the pupil part, by adjusting the brightness, contrast, etc. in the image acquisition board (image capture board), eye images having different brightnesses are obtained. It is characterized by acquiring.

In the imitation iris discrimination method of the iris recognition system of the present invention, an eye image having different brightness is obtained by adjusting the brightness of the lighting device itself to obtain a degree of change in brightness of the pupil part.

In addition, in the imitation iris discrimination method of the iris recognition system of the present invention, eye images of different brightnesses are acquired at different time intervals for measuring the brightness variation, and global illumination by XY motion correction and both lights (fire) It is characterized by performing a change correction.

In the imitation iris discrimination method of the iris recognition system of the present invention, it is characterized in that the imitation iris is further determined by further using the edge component of the boundary portion of the pupil together with the brightness gradient when determining whether the imitation iris is present. .

In addition, in the imitation iris discrimination method of the iris recognition system of the present invention, to determine whether the imitation iris or not, the degree of influence of the light reflection by the glasses is checked, and whether or not the imitation iris is considered in consideration of the brightness variation. It is characterized by determining.

Referring to the attached iris recognition method of the present invention made as described above with reference to the accompanying drawings as follows.

In the iris recognition system according to the present invention, when the user's eye image is input using an iris image input device composed of camera equipment or the like, the pupil and the iris region are extracted from the eye image information, and based on the pupil and iris region extraction information. Then, recognition is performed by extracting a feature (Iris Code) to be used for iris recognition. If it is determined that the user is a registered user, the dummy iris discrimination algorithm of the present invention is performed to determine whether the user is a real human eye or a forged eye by printing. As a result of the imitation iris discrimination algorithm, when the image is identified as the imitation eye, the eye image is acquired again, and the entire process of recognition and the identification of the imitation eye is repeated two or three times. Output

4 is a flowchart of an iris recognition method according to the present invention, in which a step of acquiring an eye image having a change in brightness (S41), a step of acquiring a pupil and an iris image having a change in brightness (S42), and a signal of the acquired iris image Processing step S43, acquiring a difference image pattern of the pupil part according to the brightness change (S44), and determining the imitation iris based on the degree of brightness variation of the pupil part, and performing recognition, registration, and determination of the iris. It is made to the step (S45).

5 illustrates the concept of an iris recognizer according to an embodiment of the present invention. As shown in FIG. 5, two recognition lighting devices R1 and R2 are installed on both sides of the camera for iris recognition. The lighting apparatuses R1 and R2 may be arranged on the left and right sides of the camera, and may be arranged above and below the camera, and one or three or more lighting apparatuses may be used.

In the present invention, the eye device is driven to obtain an eye image, and pupil and iris images are acquired from the eye image to determine a dummy iris based on the degree of change in brightness of the pupil part.

In general, the human pupil is covered with a cornea containing the lens (Cyrstaline Lens) and aqueous humor (Aqueous Humor) in front of it, but inside it is a kind of black hole (hole) have. Therefore, even when adjusting the brightness of the image input from the camera, the degree of change in brightness of the pupil portion does not appear much. Therefore, it is possible to distinguish the real human eye from the fake eye based on this property.

In the present invention, the difference image is used to obtain the degree of change in brightness of the pupil part. That is, the brightness variation of the pupil part is measured by using two eye images (a pupil and an iris image) photographed with a predetermined parallax, and when the brightness variation is larger than a predetermined threshold set for iris authenticity determination. It is judged to be an imitation iris, otherwise it is judged by the real human eye.

There may be various methods for obtaining the brightness gradient of the eye image. One method is to drive the lighting devices R1 and R2 with different brightness, acquire two eye images with a predetermined parallax, and measure the degree of change in brightness of the pupil part from the difference image of the two eye images. Another method is to obtain two eye images of different brightness by adjusting the brightness, contrast, and shutter speed on the camera itself, and to calculate the brightness variation of the pupil part from the difference image of the image. How to measure. Another method is to obtain two eye images of different brightness by adjusting the brightness, contrast, etc. in the image capture board of the image acquired by the camera, and measure the degree of change in the brightness of the pupil part from the difference image of the image. .

6 is a view showing an example of an iris image for explaining the case of the non-wearing glasses in the iris recognition method according to an embodiment of the present invention, Figure 7 shows an example of a dummy iris image in the iris recognition method according to an embodiment of the present invention Drawing.

In Fig. 6, reference numeral 61 denotes an eye image obtained with a relatively high brightness, 62 denotes an eye image obtained with a relatively low normal brightness, and 63 denotes a difference image of the two eye images. In FIG. 7, 71 is an eye image of a pseudo eye obtained at a relatively high brightness, 72 is a eye image of a simulated eye obtained at a relatively low normal brightness, and 73 is a It is the difference image of the eye image.

As can be seen in Figures 6 and 7, it can be seen that the difference in the change of the pupil due to the change in brightness shows a marked difference in the real eye and the fake eye. In other words, in the real human eye, the degree of change of pupil due to the change of brightness is relatively small, and the imitation iris can be screened out based on the degree of change of brightness using the geometrical characteristics of the pupil.

Here, the difference image may be obtained using various image processing techniques such as a gradient difference image as well as a simple image difference image.

On the other hand, as described above, in order to determine the imitation iris based on the change in the pupil, that is, the brightness change according to the brightness change, two eye images are obtained with different parallaxes. At this time, XY motion correction and global illumination change correction by both illuminations are performed. As shown in 61 and 62 of FIG. (The iris is moving) and there is a difference in brightness around the iris caused by the lighting. In the XY motion correction method, an iris center (or pupil center) is obtained for a first iris image, an iris center (or pupil center) is obtained for a second iris image, and then a difference of reference coordinate positions between the two images is obtained. Correction can be performed by moving the iris image to perform a match (match) with each other.

That is, when two consecutive images are acquired, there is an image acquisition time interval between the two images, so that the two images are matched using an image processing method, where histogram processing and dispersion are performed for fast and accurate matching between the two images. Various techniques can be used, such as an analysis of variation (N-Step Searching) and N-Step Searching.

Global illumination correction is a technique that eliminates the difference in brightness around the iris caused by the lighting. For example, a spectacle wearer may have an excessive brightness difference due to the reflection of the light on the glasses. It is used as iris recognition after correction processing by considering it as spot).

As described above, when the iris image is obtained while the motion and light changes are corrected, a more accurate three-dimensional pattern of the iris and eyebrows and the iris and iris boundary can be obtained. The iris image obtained in this manner (actual The pupil and iris patterns according to the change in the brightness of the human eye) can be clearly distinguished from the two-dimensional iris patterns shown in pictures or photographs, and thus the detection of counterfeit irises based on this can be performed with high reliability.

FIG. 8 shows a phenomenon in which the pupil changes according to the brightness change similar to the fake eye when the specular reflection is generated by the glasses in the case of the spectacle wearer. In FIG. 8, reference numeral 81 denotes an eye image of the spectacle wearer obtained at a relatively high brightness, 82 is an eye image of the spectacle wearer obtained at a relatively low normal brightness, and 83 is an image of the two eyes. It is the difference image.

As can be seen in FIG. 8, in the case of the wearer of the glasses, the change of the pupil due to the change of brightness appears similar to the fake eye under the influence of the light reflection by the glasses. You can further sift through the imitation iris by using this relatively strong character.

In addition, if the image is acquired by setting the brightness, contrast, shutter speed, etc. of the camera or the image acquisition board to a low value, the overall image brightness becomes dark, but the reflection of the illumination by the glasses remains, and thus the input image is additionally used. Find out the effect of the specular light reflection on Essau, and based on this, you can distinguish the real eye with the fake eye as described earlier.

In the present invention, when the dummy iris is discriminated in the iris recognition system, it is determined whether the dummy iris is based on the pupil change according to the brightness change of the eye image. Therefore, it is possible to reduce the time taken to determine the imitation iris, and to perform the iris recognition near the real time, and to provide more reliable imitation iris identification by applying each feature information including the brightness variation of the pupil to the same judgment step. To make it possible.

Claims (10)

Acquiring pupil and iris images from the photographed eye image, measuring brightness variation of the pupil portion of the acquired pupil and iris image, and determining whether the dummy iris is based on the measured brightness variation of the pupil portion Imitation iris identification method of the iris recognition system, characterized in that consisting of steps. The iris recognition according to claim 1, wherein the pupil part is determined to be a real human eye when the brightness variation is less than or equal to a set threshold, and the iris recognition is determined to be a case where the brightness variation of the pupil part is greater than or equal to a set threshold. How to determine the fake iris in the system. 2. The method of claim 1, wherein two eye images photographed at different times by two lighting means are used to obtain a degree of change in brightness of the pupil part. The method of claim 1, wherein the difference image of two eye images taken at two different time intervals is used to obtain a degree of change in brightness of the pupil part. The dummy iris of the iris recognition system according to claim 1, wherein an eye image having different brightness is obtained by adjusting brightness, contrast, shutter speed, and the like in the camera itself to obtain a degree of change in brightness of the pupil part. Discrimination method. The iris recognition system according to claim 1, wherein eye images having different brightnesses are acquired by adjusting brightness, contrast, and the like in an image acquisition board (image capture board) to obtain a degree of change in brightness of the pupil part. Method of discriminating imitation iris 2. The method of claim 1, wherein the eye image having different brightness is obtained by adjusting the brightness of the lighting device itself to obtain the degree of change in brightness of the pupil part. The iris of claim 1, wherein eye images having different brightnesses are acquired at different time intervals to measure the brightness gradient, and XY motion correction and global illumination variation correction by both illuminations (fire) are performed. Imitation Identification Method in Recognition System. The method of claim 1, wherein the method further comprises using the edge component of the boundary portion of the pupil to determine whether the dummy iris is included in the iris recognition system. . 2. The iris recognition system according to claim 1, wherein the degree of influence of the light reflection by the glasses is checked to determine whether the dummy iris is present, and whether the dummy iris is determined in consideration of the information along with the brightness variation. Method of discriminating imitation iris

Images