JP2006343995A - Biological information registration device or collation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a biological information registration device which acquires a biological image, confirms the image quality of the biological image thus acquired, determines whether a biological image satisfying a prescribed image quality is to be registered, and extracts and registers the characteristics of the biological image which is determined to be registered, the device preventing a biological image photographed without any permission from being fraudulently used when a registration manager checks the biological image. <P>SOLUTION: An image-dividing means 3-2 divides a biological image satisfying a prescribed image quality into partial images based on a prescribed rule. A display-part determination means 3-3 is used to determine whether the images are to be registered of from among the divided partial images divided by the image-dividing means 3-2. A registration propriety determination result accepting means 3-4 accepts the registration propriety determination result on the partial image determined by the display-part determination means 3-3. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a biometric information registration device or collation device such as an iris or a fingerprint, and more particularly, to a biometric information registration device or collation device having a function of preventing biometric information voyeurism.

  In recent years, personal authentication systems using biological information such as irises and fingerprints are being adopted in various fields. Along with this, fraudulent acts based on sneak shots, plagiarism, forgery, etc. of biometric information are increasing. These measures to prevent fraud are a major technical challenge. Technical development for improving algorithms for solving such technical problems and improving the accuracy of biological information is also in progress (Patent Document 1). However, forgery of biological information has become more sophisticated, and it has become difficult to achieve the object only by improving algorithms and improving the accuracy of biological information.

  On the other hand, due to the widespread use of special contact lenses, biometric images that are not suitable for registration are increasingly being acquired. Therefore, in a normal personal authentication system, when biometric information is registered, a biometric image satisfying a predetermined image quality is displayed on the computer screen, and a professional administrator visually determines whether or not the biometric image can be registered. In addition, registration of biological images and the like in which special contact lenses are worn is excluded.

In such visual determination, the acquired biometric image is enlarged and displayed as it is in order to make it easier for the registration manager to check the biometric image. Therefore, there is a risk that the biological image is voyeurized by a voyeur by a camera or the like. Further, when the image is left on the computer screen for some reason, there is a risk that the digital data itself of the image is stolen by an unauthorized user.
JP 2004-362619 A

  The problem to be solved is that, in the conventional visual determination, the photographed image is enlarged and displayed as it is. Therefore, there is a risk that the biometric image is voyeurized by a voyeur by a camera or the like, and if the image is left on the computer screen for some reason, There is a risk that the digital data itself may be stolen.

The first invention acquires a biometric image, confirms the image quality of the acquired biometric image, determines whether or not to register a biometric image that satisfies a predetermined image quality, An apparatus for registering biometric information for extracting and registering features of the biometric image, the image segmentation means for segmenting the biometric image satisfying the predetermined image quality into partial images based on a predetermined rule, and the image segmentation Display part determining means for determining a partial image to be used for determining whether or not to register the image from among the partial images divided by the means; And a determination result receiving means.
According to a second aspect of the present invention, a biometric image is acquired, the image quality of the acquired biometric image is confirmed, the biometric image that satisfies the predetermined image quality is determined to be collated, and the biometric image that can be collated A biometric information collating device that performs personal authentication, wherein an image dividing unit that divides a biometric image satisfying the predetermined image quality into partial images based on a predetermined rule, and a partial image divided by the image dividing unit A display portion determining means for determining a partial image for use in determining whether or not the image can be verified; and a verification availability determination result receiving means for receiving the verification result for the partial image determined by the display portion determining means. It is characterized by that.

  When the registration manager checks the biometric image, the captured image is not displayed as it is, and only the partial image divided by the image dividing unit based on a predetermined rule is displayed. Therefore, even if a voyeur is taken, there is an effect that the risk of fraudulent acts is greatly reduced only by the sneak shot of the biometric image. Further, since the display part determining means determines a partial image for use in determining whether or not to register an image based on the result of random number calculation, the display location is not specified, and the risk of further fraudulent acts being reduced. Get.

  A CPU (central processing unit) that controls the entire apparatus including all of the image dividing unit, the display portion determining unit, and the registration availability determination result receiving unit is provided in advance with a predetermined ROM (read-only memory). The computer control means is activated by executing a predetermined control program stored in (1).

FIG. 1 is a configuration block diagram of a biometric information registration apparatus according to the first embodiment.
As shown in the figure, the biometric information registration apparatus 100 according to the first embodiment includes an image acquisition unit 1, an operation unit 2, a quality confirmation unit 3, a feature extraction unit 4, an input / output interface unit 5, a CPU 6, a ROM 7, and the like. RAM8. In the following, biometric information will be described as an example limited to an iris.

  The image acquisition unit 1 is a part that acquires an iris image from a registrant, and has a living body imaging unit 1-1 and a mechanism control unit 1-2 therein. The living body imaging unit 1-1 has a camera and is a unit that images the registrant's iris based on a predetermined illuminance. The mechanism control unit 1-2 is a unit for correcting the position and orientation of the registrant's eyes, the illuminance of the illuminating lamp, and the like to obtain a predetermined image quality. The mechanism control unit 1-2 is feedback-controlled by the quality confirmation unit 3 based on the determination result by the registration manager. Of course, the registrant applicant may control it by himself.

  The operation unit 2 is a part that shares the role of a man-machine interface between the registration manager and the biometric information registration apparatus 100, and includes an image display unit 2-1 and an input operation unit 2-2 therein. Have.

  The image display unit 2-1 is a display that displays an iris image based on an instruction from the quality confirmation unit 3. It is also means for displaying operation information, inconvenience occurrence information, and the like of the entire apparatus based on direct control of the CPU 6.

  The input operation means 2-2 is means for the registration manager to input his / her judgment result based on the image displayed on the image display means 2-1 and notify the quality confirmation unit 3 of the input operation means 2-2. A switch or the like is used. Also, it is a means for allowing the registration manager to directly handle the operation of the entire apparatus, inconvenience occurrence information, etc., and allowing the registration manager to directly control the CPU 6.

  The quality confirmation unit 3 displays the iris image acquired by the image acquisition unit 1 or the data related to the image on the image display unit 2-1, causes the registration manager to confirm the image quality, and further, the predetermined image For an iris image that satisfies the quality, it is a part that determines whether or not the iris image may be registered as biometric information and accepts the result of the determination. The image quality acceptance / rejection acceptance means 3-1, image division means 3-2, display portion determination means 3-3, and registration availability judgment acceptance means 3-4 are included therein.

  The image quality acceptance / reception unit 3-1 displays the iris image acquired by the image acquisition unit 1 or data related to the image on the image display unit 2-1, and the input operation unit 2 determines whether the registration manager accepts the image quality. It is a means to accept via -2. Here, the image quality is, for example, the size of the image (registrant's eye position), the direction of the eye, the degree of eyelid opening, the brightness of the screen (illuminance of the illumination lamp), the size of the reflected light, etc. The quality state of whether or not it can be a target image for determining whether or not to register an iris image that progresses thereafter. Here, it is not always necessary to display the acquired iris image on the image display means 2-1, and it may be performed only by checking data regarding the image. Even when the acquired iris image is displayed on the image display means 2-1, the check of the above items does not require an image that is so precise and enlarged that the features of the iris image appear. Therefore, in the present invention, it is not considered as an object of voyeurism.

The image dividing unit 3-2 is a unit that divides an iris image satisfying a predetermined image quality into partial images based on a predetermined rule.
FIG. 2 is an explanatory diagram of the image dividing means.
(A) is an overall image of an iris image, and (b) is a diagram showing an example of a division rule.

  As shown in the figure, the image dividing unit 3-2 is a unit that divides the whole image (a) of the iris image into four (an example) as shown in (b). Displaying the entire iris image 10a on the screen increases the possibility that the iris image is voyeurized and used for fraud. Reference numerals 10a-1 to 10a-4 represent cases where the iris image is divided into four partial images. Here, it is defined as the partial image using four rectangles clockwise around the iris center 10a-5 that is the center of the iris (the predetermined rule).

  In the case of an iris image, such processing can cut out an appropriate rectangle because information such as the iris center 10a-5 and the iris radius is obtained in the process of detecting the iris portion. In the case of a fingerprint image or the like, an appropriate part can be defined from various information obtained in the process of acquiring the biological image. Further, whether or not the acquired iris image can be registered is determined based on an empirical rule that it is possible to visually check a part of the iris image.

  Returning to FIG. 1, the display part determining means 3-3 is a means for determining a partial image for use in determining whether or not to register an image from among the partial images divided by the image dividing means 3-2. This determination is determined using a random number. Here, as an example, one is selected from four partial images. However, it is not always necessary to follow this example. For example, two of four partial images may be selected.

  The registration availability determination accepting means 3-4 displays the partial image determined by the display part determination means 3-3 on the image display means 2-1, and makes the registration manager determine registration availability based on visual determination. It is a means for obtaining and receiving the determination result. Here, an example of registration permission / inhibition determination based on visual determination will be described.

FIG. 3 is an explanatory diagram of a registration availability determination accepting unit.
(A) represents the whole image of the iris image, (b) represents the partial image (block No. 4), and (c) represents the partial image (block No. 2). Either one is displayed during actual operation.

  As shown in (b) and (c), in this figure, a gear-like striped pattern is seen at the boundary between the iris and the pupil. From this striped pattern, it is determined that the registration applicant is wearing a color contact based on the rule of thumb of the registration manager. In such a case, since it is not suitable as a registered image, the registration manager determines that the registration is not permitted, and enters the registration permission determination acceptance means 3-4 via the input operation means 2-2 (FIG. 1). Notify the decision.

Returning to FIG. 1, the feature extraction unit 4 confirms the image quality of the iris image, determines whether or not to register an iris image satisfying a predetermined image quality, and for the iris image that can be registered, the iris image This is a part for extracting features (biological information).
The input / output interface unit 5 is a part that transmits and registers the features (biological information) of the iris image extracted by the feature extraction unit 4 to the database 50 via the network 40.

  The CPU 6 is a central processing unit that controls the entire apparatus. In particular, in the present embodiment, the image acquisition unit 1, the quality confirmation unit 3, the feature extraction unit 4, and the input / output interface unit 5 are activated as computer control means by executing a control program stored in advance in the ROM 7. It is.

  The ROM 7 executes a control program and control data for controlling the entire apparatus when executed by the CPU 6, and the image acquisition unit 1, the quality confirmation unit 3, the feature extraction unit 4, and the input / output interface unit 5 when executed by the CPU 6. Is a read-only memory that stores in advance a control program and control data for activating.

The RAM 8 is a random access memory that provides a calculation area necessary for the CPU 6 to execute the various control programs.
The common bus 9 is a signal path that connects each part.

Next, the operation of the biometric information registration apparatus according to the first embodiment will be described.
FIG. 4 is a flowchart illustrating the operation of the first embodiment.
Steps S1-1 to S1-10 will be described in the order of steps.

Step S1-1
The biological imaging apparatus (image acquisition unit 1 (FIG. 1)) is initialized and becomes operable.
Step S1-2
The image acquisition unit 1 (FIG. 1) captures an iris image (biological information).

Step S1-3
The image quality acceptance / rejection acceptance unit 3-1 (FIG. 1) accepts the iris image acquired by the image acquisition unit 1 (FIG. 1) and sends it to the registration manager via the image display unit 2-1 (FIG. 1). To obtain an input for pass / fail judgment of image quality. Here, the image quality is, for example, the size of the image (registrant's eye position), the direction of the eye, the eye opening, the brightness of the screen (illuminance of the illumination lamp), and the size of the reflected light. The quality state of whether or not it can be a target for determining whether or not to register an iris image that progresses thereafter. Here, it is not always necessary to display the acquired iris image on the image display means 2-1 (FIG. 1), and it may be performed only by data check. Even when the acquired iris image is displayed on the image display means 2-1 (FIG. 1), an image as precise as the feature of the iris image does not need to be checked for the above items.

Step S1-4
When the registration manager inputs an image quality rejection determination via the input operation means 2-2 (FIG. 1), the image quality acceptance / rejection means 3-1 (FIG. 1) is connected to the image acquisition unit 1 (FIG. 1). In contrast, step S1-2 and step S1-3 are repeatedly executed. When the registration manager inputs an image quality acceptance determination via the input operation means 2-2 (FIG. 1), the image quality acceptance / rejection acceptance means 3-1 (FIG. 1) advances the flow to step S1-5. .

Step S1-5
The image dividing means 3-2 (FIG. 1) divides the iris image determined to be accepted in step S1-4 into partial images based on a predetermined rule. Here, as shown in FIG. 2, the image is divided into four 10a-1 to 10a-3 as partial images using four rectangles clockwise around the iris center 10a-5 as the center of the iris. Subsequently, the display part determining means 3-3 (FIG. 1) selects one of the four partial images. Here, it is assumed that the partial image (block No. 2) in FIG. 3 is selected based on the calculation result of the random number calculation.

Step S1-6
The registration permission determination acceptance unit 3-4 (FIG. 1) instructs the image display unit 2-1 (FIG. 1) to cut out the partial image (block No. 2) of FIG.

Step S1-7
The registration permission determination acceptance means 3-4 (FIG. 1) instructs the image display means 2-1 (FIG. 1) to enlarge and display the partial image (block No. 2) of FIG. 3 to the registration manager. It is determined whether or not the iris image can be registered.

Step S1-8
The registration manager visually checks the partial image (block No. 2) displayed on the image display means 2-1 (FIG. 1) and determines whether registration is possible. As an example, as shown in FIG. 3C, it is assumed that a gear-like striped pattern is seen at the boundary between the iris and the pupil. From this striped pattern, based on the rule of thumb of the registration manager, it is determined that the registrant is wearing a color contact, it is determined that registration is not allowed, and the flow proceeds to step S1-9. If such a problem cannot be detected, it is determined that the registration is complete, and the process proceeds to step S1-10.

Step S1-9
The image acquisition unit 1 (FIG. 1) finishes shooting the iris image and ends the flow (registration failure).
Step S1-10
The feature extraction unit 4 (FIG. 1) extracts biological information (characteristics of the iris image) from the iris image, and the input / output interface unit 5 (FIG. 1) is connected to the database 50 (FIG. 1) via the network 40 (FIG. 1). The biometric information (characteristic of the iris image) is registered and the flow ends (registration success).

  As described above, in this embodiment, when the registration manager checks the biometric information, the captured image is not displayed in the form as it is, but is divided based on a predetermined rule by the image dividing means. Since only the partial image that has been displayed is displayed, there is an effect that the risk of fraudulent acts due to voyeurism or the like is greatly reduced. Further, since the display part determining means determines a partial image for use in determining whether or not to register an image based on the result of random number calculation, the display location is not specified, and the risk of further fraud is reduced. Get.

  In the above description, the biological information is limited to only the iris, but the present invention is not limited to this example. That is, the present invention can also be applied to other biometric information registration devices such as fingerprints. In addition, the operations from step S1-1 to step S1-8 in the first embodiment can be applied not only to the biometric information registration apparatus but also to the collation apparatus. However, when it is applied to a collation device, the iris image is stored as history information, and a slight change in content is involved, such as regular determination.

FIG. 5 is a configuration block diagram of the biometric information registration apparatus according to the second embodiment.
As shown in the figure, the biometric information registration apparatus 200 according to the second embodiment includes an image acquisition unit 1, an operation unit 2, a quality confirmation unit 21, a feature extraction unit 4, an input / output interface unit 5, a CPU 22, a ROM 23, and the like. RAM8. Only differences from the first embodiment will be described below. The same parts as those in the first embodiment are denoted by the same reference numerals as those in the first embodiment, and the description thereof is omitted.

  The quality confirmation unit 21 displays the iris image acquired by the image acquisition unit 1 or data related to the image on the image display unit 2-1, causes the registration manager to confirm the image quality, and obtains a predetermined image quality. For a satisfactory iris image, it is a part that obtains a determination as to whether or not the iris image may be registered as biometric information and accepts the determination result. In the image quality acceptance / rejection acceptance means 3-1, image division means 3-2, display portion determination means 3-3, registration availability judgment acceptance means 3-4, and registration inappropriateness emphasis means 21-1 are provided. Have.

The registration inappropriateness emphasis emphasizing means 21-1 detects a portion having a possibility of inappropriate registration by performing software processing on the iris image acquired by the image acquisition unit 1, emphasizes the registration inappropriate location, and the registration manager This is a means for facilitating visual judgment. For example, it is executed when the following phenomenon is detected by software processing.
(1) Despite the fact that the human pupil is supposed to be black because it absorbs light, there are places where a line (straight line or curve) extending from the iris part protrudes from the pupil part. If
(2) When there is a long continuous line segment with the same brightness extending from the inside to the outside of the iris,
(3) This is the case when spot patterns with the same interval are regularly arranged inside the iris.
When such a phenomenon is detected by software processing, a pattern portion different from the human eye is colored with a different color.

FIG. 6 is an explanatory diagram of the registration inappropriateness possibility emphasizing means.
(A) represents the case where the color contact is a striped pattern, and the striped pattern is emphasized by coloring in blue. (B) shows the case where the color contact has a spotted pattern, and the spotted pattern is emphasized by coloring in red.

  The CPU 22 is a central processing unit that controls the entire apparatus. In particular, in the present embodiment, the image acquisition unit 1, the quality confirmation unit 21, the feature extraction unit 4, and the input / output interface unit 5 are activated as computer control means by executing a control program stored in advance in the ROM 23. It is.

The ROM 23 executes a control program and control data for controlling the entire apparatus when executed by the CPU 22, and the image acquisition unit 1, the quality confirmation unit 21, the feature extraction unit 4, and the input / output interface unit 5 when executed by the CPU 22. Is a read-only memory that stores in advance a control program and control data for activating.
The other parts are the same as those in the first embodiment, and the description thereof is omitted.

Next, the operation of the biometric information registration apparatus according to the second embodiment will be described.
FIG. 7 is a flowchart illustrating the operation of the second embodiment.
Steps S2-1 to S2-11 will be described in the order of steps.

Step S2-1
The biological imaging apparatus (image acquisition unit 1 (FIG. 5)) is initialized and becomes operable.
Step S2-2
The image acquisition unit 1 (FIG. 5) captures an iris image (biological information).

Step S2-3
The image quality acceptance / rejection acceptance unit 3-1 (FIG. 5) accepts the iris image acquired by the image acquisition unit 1 (FIG. 5) and notifies the registration manager via the image display unit 2-1 (FIG. 5). To obtain an input for pass / fail judgment of image quality. Here, the image quality is, for example, the size of the image (registrant's eye position), the direction of the eye, the eyelid opening, the brightness of the screen (illuminance of the illumination lamp), and the size of the reflected light. The quality state of whether or not it can be a target for determining whether or not to register an iris image that progresses thereafter. Here, it is not always necessary to display the acquired iris image on the image display means 2-1 (FIG. 5). Even when the acquired iris image is displayed on the image display means 2-1 (FIG. 5), an image as precise as the feature of the iris image does not need to be checked for the above items.

Step S2-4
When the registration manager inputs an image quality rejection determination via the input operation unit 2-2 (FIG. 5), the image quality acceptance / rejection unit 3-1 (FIG. 5) Steps S2-2 and S2-3 are repeatedly executed. When the registration manager inputs an image quality acceptance determination via the input operation means 2-2 (FIG. 5), the image quality acceptance / rejection acceptance means 3-1 (FIG. 5) advances the flow to step S2-5. .

Step S2-5
The image dividing unit 3-2 (FIG. 5) divides the iris image determined to be accepted in step S 2-4 into partial images based on a predetermined rule. Here, as shown in FIG. 2, the image is divided into four 10a-1 to 10a-3 as partial images using four rectangles clockwise around the iris center 10a-5 as the center of the iris. Subsequently, the display part determining unit 3-3 (FIG. 5) selects one of the four partial images. Here, it is assumed that the partial image (block No. 2) in FIG. 3 is selected based on the calculation result of the random number calculation.

Step S2-6
The registration permission determination acceptance means 3-4 (FIG. 5) instructs the image display means 2-1 (FIG. 5) to cut out the partial image (block No. 2) of FIG.

Step S2-7
The registration inappropriateness emphasis emphasizing means 21-1 (FIG. 5) detects a portion that may be inappropriate for registration by performing software processing on the iris image acquired by the image acquisition unit 1 (FIG. 5). Emphasize locations that are not suitable for registration, and make it easier for the registration manager to visually check.

Step S2-8
The registration permission determination acceptance unit 3-4 (FIG. 5) instructs the image display unit 2-1 (FIG. 5) to enlarge and display the partial image (block No. 2) of FIG. 3 to the registration manager. It is determined whether or not the iris image can be registered.

Step S2-9
The registration manager visually checks the partial image (block No. 2) displayed on the image display means 2-1 (FIG. 5) and determines whether registration is possible. As an example, as shown in FIG. 6A, the stripe pattern is emphasized by coloring in blue. If it is determined from this striped pattern that the registrant is wearing a color contact based on the rule of thumb of the registration manager, it is determined that the registration is rejected and the process proceeds to step S2-10. If such a problem cannot be detected, it is determined that the registration is complete, and the process proceeds to step S2-11.

Step S2-10
The image acquisition unit 1 (FIG. 5) finishes shooting the iris image and ends the flow (registration failure).
Step S2-11
The feature extraction unit 4 (FIG. 5) extracts biological information (features of the iris image) from the iris image, and the input / output interface unit 5 (FIG. 5) is connected to the database 50 (FIG. 5) via the network 40 (FIG. 5). The biometric information (characteristic of the iris image) is registered and the flow ends (registration success).

  As described above, in this embodiment, since the registration inappropriateness emphasis means is provided and the registration inappropriateness possibility portion is emphasized, in addition to the effects of the first embodiment, the judgment of the registration manager is facilitated. The effect that it can be obtained. In addition, the registration inappropriateness emphasis means merely emphasizes the location where registration is not possible, and the judgment is only executed by the registration manager. However, it can be corrected by the registration manager.

  In the above description, the biological information is limited to only the iris, but the present invention is not limited to this example. That is, the present invention can be applied in the same manner to other biometric information registration devices such as fingerprints. Further, the operations from step S2-1 to step S2-9 in the second embodiment can be applied to a biometric information collating device as it is. However, when it is applied to a collation device, the iris image is stored as history information, and a slight change in content is involved, such as regular determination.

FIG. 8 is a configuration block diagram of the biometric information registration apparatus according to the third embodiment.
As shown in the figure, the biometric information registration device 300 according to the third embodiment includes an image acquisition unit 1, an operation unit 2, a quality confirmation unit 31, a feature extraction unit 4, an input / output interface unit 5, a CPU 32, and a ROM 33. RAM8. Only differences from the first embodiment or the second embodiment will be described below. Parts similar to those in the first or second embodiment are denoted by the same reference numerals as those in the first or second embodiment, and the description thereof is omitted.

  The quality confirmation unit 31 displays the iris image acquired by the image acquisition unit 1 or data related to the image on the image display unit 2-1, causes the registration manager to confirm the image quality, and obtains a predetermined image quality. For a satisfactory iris image, it is a part that obtains a determination as to whether or not the iris image may be registered as biometric information and accepts the determination result. In the image quality acceptance / rejection acceptance means 3-1, image division means 3-2, display portion determination means 3-3, registration availability judgment acceptance means 3-4, and registration inappropriateness emphasis enhancement means 21-1. The display portion position specifying means 31-1 is provided.

The display part position specifying means 31-1 is a means for instructing the image display means 2-1 as to which position of the eye the partial image magnified and displayed as the display part corresponds to as an image.
FIG. 9 is an explanatory diagram of the display portion position specifying means.
(A) is an image displayed on the image display means 2-1 by the display portion position specifying means 31-1, and (b) is a registration availability determination accepting means as in the first or second embodiment. The image is displayed on the image display unit 2-1 by 3-4. Both are displayed on the same screen. With these displays, it becomes clear to which position of the eye the partial image displayed on the image display means 2-1 by the registration permission determination acceptance means 3-4.

  Returning to FIG. 8, the CPU 32 is a central processing unit that controls the entire apparatus. In particular, in the present embodiment, the image acquisition unit 1, the quality confirmation unit 31, the feature extraction unit 4, and the input / output interface unit 5 are activated as computer control means by executing a control program stored in advance in the ROM 33. It is.

The ROM 33 executes a control program and control data for controlling the entire apparatus when executed by the CPU 32, and the image acquisition unit 1, the quality confirmation unit 31, the feature extraction unit 4, and the input / output interface unit 5 when executed by the CPU 32. Is a read-only memory that stores in advance a control program and control data for activating.
The other parts are the same as those in the first embodiment or the second embodiment, and thus the description thereof is omitted.

Next, the operation of the biometric information registration apparatus according to the third embodiment will be described.
FIG. 10 is a flowchart illustrating the operation of the third embodiment.
Steps S3-1 to S3-12 will be described in the order of steps.

Step S3-1
The biological imaging apparatus (image acquisition unit 1 (FIG. 8)) is initialized and becomes operable.
Step S3-2
The image acquisition unit 1 (FIG. 8) captures an iris image (biological information).

Step S3-3
The image quality acceptance / rejection acceptance unit 3-1 (FIG. 8) accepts the iris image acquired by the image acquisition unit 1 (FIG. 8) and sends it to the registration manager via the image display unit 2-1 (FIG. 8). To obtain an input for pass / fail judgment of image quality. Here, the image quality is, for example, the size of the image (registrant's eye position), the direction of the eye, the eye opening, the brightness of the screen (illuminance of the illumination lamp), and the size of the reflected light. The quality state of whether or not it can be a target for determining whether or not to register an iris image that progresses thereafter. Here, it is not always necessary to display the acquired iris image on the image display means 2-1 (FIG. 8). Even when the acquired iris image is displayed on the image display means 2-1 (FIG. 8), the check of the above items does not require an image that is so precise that the feature of the iris image appears.

Step S3-4
When the registration manager inputs an image quality rejection determination via the input operation means 2-2 (FIG. 8), the image quality acceptance / rejection acceptance means 3-1 (FIG. 8) displays the image acquisition unit 1 (FIG. 8). In contrast, step S3-2 and step S3-3 are repeatedly executed. When the registration manager inputs an image quality acceptance determination via the input operation means 2-2 (FIG. 8), the image quality acceptance / rejection acceptance means 3-1 (FIG. 8) advances the flow to step S3-5. .

Step S3-5
The image dividing unit 3-2 (FIG. 8) divides the iris image determined to be acceptable in step S3-4 into partial images based on a predetermined rule. Here, as shown in FIG. 2, the image is divided into four 10a-1 to 10a-3 as partial images using four rectangles clockwise around the iris center 10a-5 as the center of the iris. Subsequently, the display part determining means 3-3 (FIG. 8) selects one of the four partial images. Here, it is assumed that the partial image (block No. 2) in FIG. 3 is selected based on the calculation result of the random number calculation.

Step S3-6
The registration permission determination acceptance unit 3-4 (FIG. 8) instructs the image display unit 2-1 (FIG. 8) to cut out the partial image (block No. 2) of FIG.

Step S3-7
The display part position specifying means 31-1 instructs the image display means 2-1 as to which position of the eye the partial image determined as the display part corresponds to (FIG. 9 (a). )).

Step S3-8
The registration inappropriateness emphasizing means 21-1 (FIG. 8) detects a portion that may be inappropriate for registration by performing software processing on the iris image acquired by the image acquisition unit 1 (FIG. 8), and the registration inappropriateness is possible. Emphasize the sex location and make it easier for the registration manager to visually check.

Step S3-9
The registration permission determination acceptance means 3-4 (FIG. 8) instructs the image display means 2-1 (FIG. 8) to enlarge and display the partial image (block No. 2) of FIG. 3 to the registration manager. It is determined whether or not the iris image can be registered.

Step S3-10
The registration manager visually checks the partial image (block No. 2) displayed on the image display means 2-1 (FIG. 8) and determines whether registration is possible. As an example, as shown in FIG. 6A, the stripe pattern is emphasized by coloring in blue. If it is determined from this striped pattern that the registrant is wearing a color contact based on the rule of thumb of the registration manager, it is determined that the registration is not permitted and the process proceeds to step S3-11. If such a problem cannot be detected, it is determined that the registration is complete, and the process proceeds to step S3-12.

Step S3-11
The image acquisition unit 1 (FIG. 8) finishes shooting the iris image and ends the flow (registration failure).
Step S3-12
The feature extraction unit 4 (FIG. 8) extracts biological information (features of the iris image) from the iris image, and the input / output interface unit 5 (FIG. 8) is connected to the database 50 (FIG. 8) via the network 40 (FIG. 8). The biometric information (characteristic of the iris image) is registered and the flow ends (registration success).

  As described above, in the present embodiment, in addition to the second embodiment, since the portion of the eye that is likely to be inappropriate for registration is visually displayed, the effect of the second embodiment is achieved. In addition, when the registration manager determines that there is a possibility that the registration is inappropriate, an effect of facilitating the confirmation work when actually confirming the eyes of the registration applicant can be obtained.

  In the above description, the biological information is limited to the iris only, but the present invention is not limited to this example. That is, the present invention can also be applied to other biometric information registration devices such as fingerprints. Further, the operations from step S3-1 to step S3-9 in the second embodiment can be applied to a biometric information collating device. However, when it is applied to a collation device, the iris image is stored as history information, and a slight change in content is involved, such as regular determination.

  In the above description, only the case where the present invention is applied to a biometric information registration apparatus or a collation apparatus that uses iris information has been described, but the present invention is not limited to this example. That is, the present invention can be similarly applied to a biometric information registration apparatus or a collation apparatus that uses other biometric information such as a fingerprint.

1 is a configuration block diagram of a biometric information registration device according to Embodiment 1. FIG. It is explanatory drawing of an image division means. It is explanatory drawing of a registration availability determination acceptance means. 3 is a flowchart showing the operation of the first embodiment. It is a block diagram of the configuration of the biometric information registration apparatus according to the second embodiment. It is explanatory drawing of a registration inappropriateness possibility emphasis means. 10 is a flowchart showing the operation of the second embodiment. It is a block diagram of the configuration of the biometric information registration apparatus according to the third embodiment. It is explanatory drawing of the position explicit means of a display part. 10 is a flowchart illustrating the operation of the third embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image acquisition part 1-1 Living body imaging | photography means 1-2 Mechanism control means 2 Operation part 2-1 Image display means 2-2 Input operation means 3 Quality confirmation part 3-1 Image quality acceptance / reception means 3-2 Image division means 3 -3 Display part determination means 3-4 Registration acceptability judgment acceptance means 4 Feature extraction part 5 Input / output interface part 6 CPU
7 ROM
8 RAM
9 Common bus

Claims (10)

The biometric image is acquired, the image quality of the acquired biometric image is confirmed, the registration of the biometric image satisfying the predetermined image quality is determined, and the characteristics of the biometric image are determined for the biometric image that can be registered. A biometric information registration device that extracts and registers,
Image dividing means for dividing the biological image satisfying the predetermined image quality into partial images based on a predetermined rule;
Display part determining means for determining a partial image for use in determining whether or not to register the image from the partial images divided by the image dividing means;
A biometric information registration apparatus comprising: a registration availability determination result acceptance unit that accepts the registration availability determination result for the partial image determined by the display portion determination unit.   2. The biometric information registration apparatus according to claim 1, wherein the display portion determination unit determines a partial image to be used for the determination of whether or not to register based on a calculation result of a random number calculation.   The image processing apparatus further comprises registration improper possibility emphasizing means for detecting a part having a possibility of inappropriate registration from the partial image determined by the display part determining means and applying a predetermined process to the part having the possibility of inappropriate registration. The biometric information registration device according to claim 1, wherein:   The biometric information registration apparatus according to claim 3, wherein the predetermined process performs a coloring process different from other parts on the part that may be inappropriate for registration.   5. The display portion position specifying means for displaying an image indicating which position of the eye the portion to be emphasized by the registration inappropriateness emphasizing means further corresponds to. Biometric information registration device. Biometric information for acquiring a biometric image, confirming the image quality of the acquired biometric image, determining whether biometric images that satisfy a predetermined image quality can be collated, and performing personal authentication on the biometric images that can be collated The verification device of
Image dividing means for dividing the biological image satisfying the predetermined image quality into partial images based on a predetermined rule;
Display part determining means for determining a partial image for use in determining whether or not to collate the image from the partial images divided by the image dividing means;
A biometric information collating apparatus, comprising: a collation permission / inhibition determination result receiving unit that receives the collation permission / inhibition determination result for the partial image determined by the display portion determination unit.   The biometric information collating apparatus according to claim 6, wherein the display part determining unit determines a partial image to be used for determining whether or not the registration is possible based on a calculation result of a random number calculation.   The image processing apparatus further includes a non-matching possibility emphasizing unit that detects a portion having a possibility of non-matching with respect to the partial image determined by the display part determining unit, and applies a predetermined process to the portion having the non-matching possibility. The biometric information collating device according to claim 6 or 7, wherein:   9. The biometric information collating apparatus according to claim 8, wherein the predetermined processing performs a coloring process different from other portions on a portion where the collation may be inappropriate. 10. The living body according to claim 8, further comprising a display portion position specifying means for displaying an image indicating which position of the eye the portion emphasized by the non-matching possibility emphasis means corresponds to. Information verification device.

Images