KR20030070183A - Iris recording and recognition method of forged distingction of iris recognition system - Google Patents

Abstract

The present invention discloses a registration and recognition method for forgery determination of an iris recognition system. A registration method for forgery determination of an iris recognition system according to the present invention includes: acquiring an iris image of a user by a camera; Generating an iris code of the obtained iris image; Registering the generated iris code in a database; After registering the iris code, applying an iris image to a forgery determination algorithm; Setting an adaptive threshold value of a user for each determination method of the forgery determination algorithm; And storing the adaptive threshold of each set determination method in a database.

In the falsification determination method of the iris recognition system according to the present invention, it is possible to accurately determine the forgery by applying a fixed threshold, which is a criterion of the forgery discrimination algorithm, as an adaptive threshold unique to each user.

Description

IRIS RECORDING AND RECOGNITION METHOD OF FORGED DISTINGCTION OF IRIS RECOGNITION SYSTEM}

The present invention relates to a registration and a recognition method for forgery determination of an iris recognition system. In particular, by determining a fixed threshold value that is a criterion for forgery determination algorithm as an adaptive threshold value unique to each user, accurate forgery determination can be performed. The present invention relates to a forgery discrimination method of an iris recognition system.

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 a high recognition rate and more accurate recognition than the fingerprint recognition. The iris recognition system is a system that authenticates an individual by comparing a characteristic pattern of an iris with an image processing technology in an iris image captured by a video camera and comparing it with previously registered iris data.

FIG. 1 is a diagram schematically illustrating a configuration of a general iris recognition system. Referring to FIG. 1, an operation of an iris recognition system is as follows.

When the user approaches the iris recognition system, the distance measuring sensor 109 measures the distance to the user, and the controller 105 receives the distance measurement value obtained through the driver 107 to operate the distance measurement value. Determine if you are in range.

If the user is within the operating range, a control signal is sent to the driver 107 to prepare to extract the iris image. The driver 107 sends an active signal to the external indicator 108 to inform the user that the system is operating. Accordingly, when the user places the eye on the optical axis of the camera 103 through the optical window 101, the cold mirror ( 102 blocks visible light and passes infrared light. In addition, by displaying the part where the iris should be positioned, the user can check whether the eye is located on the optical axis of the camera 103.

The controller 105 receives the distance measurement value from the distance measurement sensor 109 to the user, and calculates the zoom and focus values of the camera 103 using the distance measurement value to perform zoom in / zoom out and focusing control. Perform. Thereafter, the controller 105 allows the driver 107 to capture the iris image according to the distance measurement value, and the photographed iris image is signal-processed in the frame grabber 104 for iris image analysis, and the processed iris is processed. The image information is stored in the DB 110 (database), and with the stored information, the control device 105 performs iris recognition to determine whether the user is authenticated.

On the other hand, in such an iris recognition system, the performance depends on how fast and how accurately the iris recognition is performed. In addition, when the recognition is attempted using the iris image simulated by the iris recognition system, a recognition ability for determining the recognition is required.

Therefore, in order to determine forgery of iris recognition, a method of using a high frequency pattern, a method of determining whether there is reflection in the pupil, a method of using a change in wavelength band of the iris and white, a method of using the degree of contraction of the pupil, etc. There are ways.

2 is a configuration diagram of a counterfeit discrimination apparatus using general multiple illumination. As shown here, the distance measuring unit 201 for measuring the distance from the camera to the user's eyes using the infrared; A camera 202 for photographing the eyes of the person to be recognized; Discriminating illumination 203 provided in proximity to the optical axis of the camera 202; An illumination control unit 206 for lighting the recognition lights 204 and 205 installed on the left and right sides of the camera and the discriminating light 203; And an image processor 207 for processing the output image signal of the camera 202 to extract an image of the iris and the pupil part.

Herein, a method of determining whether a forgery is determined by determining the presence of reflection in the pupil among the above methods will be described.

FIG. 3 is a flowchart illustrating a forgery discriminating method according to the presence of reflection in a pupil according to the related art, and will be described with reference to FIG. As shown in FIG. 2, when the user approaches the operating area, while lighting the discriminating light emitting diode 203 installed close to the camera 202 (S301), the user's eyes are photographed to extract an iris image (S302). ).

Subsequently, a reflection image of the pupil area is detected in the extracted iris image (S303), and whether the forgery is determined according to the presence of the reflection of the pupil area (S304). Here, whether the pupil area is reflected or not is determined by the threshold value already set.

If the reflection image is not detected in the pupil area, it is determined to be a forgery (S305), and if detected, it is recognized as an actual iris (S306).

On the other hand, the method of using the wavelength band gradient of the iris and the white egg was applied by setting the threshold in the basic assumption that the human eye is not distinguishable between the white and the iris in the long wavelength band (more than 800 nm). In the case of humans, since the light reflection is a lot in the white, the level difference between the white and the iris is represented.

In addition, the method of using the degree of contraction of the pupil is to determine the degree of contraction and relaxation of the pupil in accordance with the threshold value already set. However, as the person ages, the degree of contraction and relaxation of the pupil due to external visible light becomes much smaller.

Therefore, although these things appear to a very small part of people, when these users are to be iris recognition, recognition rejection occurs by a predetermined judgment threshold.

Therefore, the above methods have a problem that it is difficult to apply the same threshold value to all users because the degree of change is different for each user.

The present invention has been created to improve the above problems, and in particular, by applying a fixed threshold, which is a criterion for determining a forgery discrimination algorithm, as an adaptive threshold unique to each user, an iris recognition system capable of accurate forgery determination can be determined. Its purpose is to provide a registration and recognition method for forgery determination.

1 is a view schematically showing the configuration of a general iris recognition system.

2 is a block diagram of a counterfeit discrimination apparatus using a general multiple illumination.

3 is a flow chart of a forgery determination method according to the presence or absence of reflection in the pupil according to the prior art.

Figure 4 is a registration method for forgery determination of the iris recognition system according to the present invention.

5 is a flowchart of a recognition method for forgery determination of an iris recognition system according to the present invention;

In order to achieve the above object, the registration method for forgery determination of the iris recognition system according to the present invention,

Acquiring an iris image of a user by a camera;

Generating an iris code of the obtained iris image;

Registering the generated iris code in a database;

After registering the iris code, applying an iris image to a forgery determination algorithm;

Setting an adaptive threshold value of a user for each determination method of the forgery determination algorithm;

It is characterized in that it comprises the step of storing in the database an adaptive threshold of each set determination method.

Here, in particular, in the step of setting an adaptive threshold of the user for each determination method of the forgery determination algorithm, the thresholds such as intra-pupillary reflection, pupil contraction and relaxation, which are criteria of determination in each determination method, are adapted to each user. Its feature is that it sets a threshold.

In addition, in order to achieve the above object, a recognition method for forgery determination of the iris recognition system according to the present invention,

Acquiring an iris image of a user by a camera;

Generating an iris code of the obtained iris image;

Retrieving the generated iris code in a database;

After retrieving the iris code, applying an iris image to a forgery determination algorithm;

Searching for an adaptive threshold value of each determination method in the database of the forgery determination algorithm, and selecting a determination method suitable for a user;

It is characterized in that it comprises the step of notifying the recognition by determining whether the forgery by the selected determination method.

According to the present invention as described above, it is possible to make accurate forgery determination by determining a fixed threshold which is a criterion of the forgery discrimination algorithm as an adaptive threshold unique to each user.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

4 is a flowchart illustrating a registration method for forgery determination of an iris recognition system according to the present invention. As shown in the figure, when the user first approaches the recognition distance, the camera acquires an iris image of the user (S401).

The obtained iris image is processed into a digital signal to generate an iris code (S402).

Subsequently, the generated iris code is registered in the database (S403).

After registering the iris code, the iris image is applied to a forgery determination algorithm (S404). Here, the forgery determination algorithm is to determine when the iris recognition is attempted using a fake eye, etc., using a high frequency pattern, a method for determining the presence of reflection in the pupil, using a change in the wavelength band of the iris and the white Various methods such as the method, the method using the degree of contraction of the pupil, and the like are applied.

Subsequently, an adaptive threshold of the user is set for each determination method of the forgery determination algorithm (S405). In this case, in particular, each determination method of the forgery determination algorithm is determined based on a threshold value, which determines the degree of change of intra-pupil reflection, pupil shrinkage and relaxation, etc., as the reference value in each determination method. In addition, since the degree of change has a different value for each user, a threshold value is set in consideration of the degree of change unique to each user. For example, a person with a small degree of pupillary change considers the same criteria at the time of registration and recognition to reduce the probability of failure of forgery.

Then, the adaptive threshold value of each set determination method is stored in the database (S406).

5 is a flowchart of a recognition method for forgery determination of an iris recognition system according to the present invention. As shown in the figure, when the user approaches the recognition distance, the camera acquires an iris image of the user (S501), and generates an iris code of the obtained iris image (S502).

Subsequently, the generated iris code is searched in a database to determine whether a user's iris code exists (S503). When the iris code is found and the iris code of the user exists, the iris code is applied to the forgery determination algorithm to determine whether the iris image is forged (S504).

Subsequently, an adaptive threshold value of each determination method is searched in the database of the forgery determination algorithm, and a determination method suitable for a user is selected (S505). In this case, the threshold value, which is a criterion for each determination method, is determined based on the same criteria as for registration. In addition, among the various forgery determination methods, each user can select and determine a method that can more accurately determine forgery. By way of example, if a method is selected in which the difference between the threshold value between registration and recognition is large, a more robust judgment can be made.

Finally, whether the forgery is determined by the selected determination method and the recognition result is notified (S506).

Although the present invention has been described with reference to the embodiments illustrated in the drawings, this is merely exemplary, and it will be understood by those skilled in the art that various modifications and equivalent other embodiments are possible. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

As described above, in the falsification determination method of the iris recognition system according to the present invention, it is possible to accurately determine the forgery by applying a fixed threshold, which is a criterion of the forgery discrimination algorithm, as an adaptive threshold unique to each user. .

Claims (3)

Acquiring an iris image of a user by a camera; Generating an iris code of the obtained iris image; Registering the generated iris code in a database; After registering the iris code, applying an iris image to a forgery determination algorithm; Setting an adaptive threshold value of a user for each determination method of the forgery determination algorithm; And storing the adaptive threshold of each set determination method in a database. The method of claim 1, In the step of setting an adaptive threshold of the user for each determination method of the forgery determination algorithm, the thresholds, such as whether the reflection within the pupil, whether the contraction and relaxation of the pupil, which is a criterion in the determination method, are set for each user A registration method for forgery determination of the iris recognition system, characterized in that. Acquiring an iris image of a user by a camera; Generating an iris code of the obtained iris image; Retrieving the generated iris code in a database; After retrieving the iris code, applying an iris image to a forgery determination algorithm; Searching for an adaptive threshold value of each determination method in the database of the forgery determination algorithm, and selecting a determination method suitable for a user; Recognizing the forgery by the selected determination method comprising the step of notifying the recognition.