JP2010244218A - Biometric authentication device and system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent that feedback information is used for a hill climbing attack for bogus living body development, even if the feedback information is displayed or output. <P>SOLUTION: In this biometric authentication device, processing of converting a verification distance value obtained by verifying a biometric feature acquired from a user and a biometric feature previously stored in a storage means into a value of a range smaller than a threshold when receiving verification, and a value of a range of the threshold or above when rejecting verification is performed. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a biometric authentication system for identifying a person using biometric information such as a fingerprint or a vein pattern.

  Biometric authentication is widely used in financial institutions and other societies for identity verification. The general process flow of biometric authentication is as follows. For example, in the case of fingerprint authentication, a fingerprint pattern is acquired as an image through a sensor or the like, a fingerprint feature amount is extracted from the image, and a fingerprint feature amount registered in advance as a principal, that is, a match or mismatch with a template The degree (collation distance value) is calculated. If the collation distance value is smaller than a preset threshold value, the person is determined to be the person, and if it is larger, the person is determined to be another person (see Patent Document 1).

  In such a biometric authentication process, “personal refusal” is generally a problem. That is, despite the biometric feature amount extracted from the person, the collation distance value with the template is larger than the threshold value, so that it is erroneously determined as another person. On the other hand, since the collation distance value with the template is smaller than the threshold value in spite of the biometric feature amount extracted from another person, it may be erroneously determined as the principal. This is called “accepting another person”, but it leads to impersonation. Therefore, the threshold is usually set so that the acceptance of another person is overwhelmingly smaller than the rejection of the person. For this reason, since the rejection of the person frequently occurs in the normal operation, the user of biometric authentication often feels uneasy about whether the living body is held over the sensor or whether the sensor fails and does not operate. Therefore, in the fingerprint authentication system for login use of personal computers and some vein authentication systems, the extracted feature amount or the image itself captured by the sensor is displayed on the screen. As a result, the user's anxiety when refusal of the person occurs is wiped out, and the user can confirm how the living body is placed in a bad manner.

JP 2008-181570 A JP 2008-46677 A

  Such extracted biometric features, biometric information (images) captured by sensors, or collation distance values are referred to as “feedback information” as information output from the biometric authentication system. This display of feedback information involves a risk of being exploited in the development of counterfeit organisms that lead to impersonation of others.

  Specifically, it is assumed that the collation distance value is abused for the development of a counterfeit organism as feedback information. First, when an arbitrary counterfeit biological body is held over the biometric authentication device, the biometric authentication device acquires an image, extracts a feature amount, compares it with a registered template, and calculates a collation distance value. As a matter of course, since the collation distance value exceeds the threshold value, collation is rejected. Next, an arbitrary improvement is applied to the forged biometric, and the verification distance value is calculated by holding it over the biometric authentication device again in the same manner as described above. As a result, if the value is smaller than the previous collation distance value, the improvement is considered to be correct, the same improvement is continued again, and if the value is increased, the improvement is considered to be wrong. To improve. This series of counterfeit biological improvement cycles is repeated many times until the collation distance value falls below the threshold value.

  Such an attack is called a hill climbing attack. In other words, the verification distance value when completing a counterfeit organism is compared to the top of a hill, and the verification distance value is gradually reduced, that is, an attack that improves the counterfeit organism so as to climb the hill. is there. This attack does not need to know the details of the biometric feature extraction algorithm or the matching algorithm at all, and can be handled as a black box.

  There is also an idea that an image of a living body or a biological feature amount may be displayed in a blurred manner. Even in such a case, it is difficult to prevent a hill climbing attack because the collation distance value can be defined by a blurred image.

  The present invention has been made to solve at least a part of the above problems. An object of the present invention is to prevent feedback information from being displayed or output from being used in a hill climbing attack for developing a counterfeit organism.

  Patent Document 2 discloses a biometric authentication device that is not abused even if biometric information is stolen. This biometric authentication device displays a part of the image that has been processed to overwrite the image of the other part, but this processed image may give the user a sense of incongruity. Therefore, an object of the present invention is to provide biometric authentication that does not give the user a sense of incongruity.

  In the biometric authentication device, the collation distance value obtained by collating the biometric feature amount acquired from the user with the biometric feature amount stored in advance in the storage means is within a range smaller than the threshold when accepting collation. When the collation is rejected, a process of converting the value into an arbitrary value within the threshold or more is performed.

  According to the present invention, it is possible to prevent a malicious third party from using a feedback information output from a biometric authentication device to create a counterfeit biometric. That is, even if you try to improve the counterfeit organism little by little to reduce the verification distance value, by changing the verification distance value arbitrarily and randomly, whether the change in the verification distance value is due to the improvement of the counterfeit organism, It is not clear whether the collation distance value is arbitrarily changed on the system side. As a result, the direction of improvement of counterfeit organisms is not understood, and development of counterfeit organisms becomes extremely difficult.

The figure which shows the example of whole structure of a biometrics authentication apparatus and a computer. The figure which shows the structural example of a biometrics authentication apparatus. The figure which shows the structural example of the computer to which the biometrics apparatus was connected. The figure which shows the operation | movement flow of a biometrics authentication apparatus and a computer. The figure which shows the example which converts collation distance value using a random number. The figure which shows transition of the collation distance value at the time of using this invention. The figure which shows the example of the biometric image for a display produced | generated by the image process of the blood vessel part. The figure which shows the example of the biological image for a display produced | generated by the image process of the blood vessel background part. The figure which shows the example of a screen display of an authentication result. The figure which shows transition of the collation distance value in the process of developing a forged living body. The figure which shows the conversion table for replacing the collation distance value of a continuous value with a discrete value.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS.

  In this embodiment, it is assumed that a biometric authentication device is connected to a computer, and feedback information generated by the biometric authentication device is sent to the computer and displayed on the computer display.

  FIG. 1 shows a configuration example of the entire system. Reference numeral 101 denotes a biometric authentication device, and reference numeral 102 denotes a computer that controls activation and operation of the biometric authentication device 101 and displays a collation result and a biometric image.

  FIG. 2 is a diagram illustrating a configuration example of the biometric authentication device 101. A control unit (CPU) 201 is a processor responsible for data processing of the biometric authentication device, and controls various programs and data described later. The peripheral device I / O device 202 is an interface for connecting the biometric authentication device 101 and the computer 102. The illumination LED 203 is an illumination LED for acquiring a biological image. For example, in the case of finger vein authentication, a near-infrared light LED suitable for acquiring a finger vein pattern is used. The image sensor 204 is a sensor for acquiring a living body image, and includes a device such as a CCD, and acquires a finger vein pattern irradiated by the illumination LED 203. A bus 205 connects devices in the biometric authentication apparatus.

  The main storage device 206 is composed of a volatile memory (DRAM or the like), in which various programs and data areas for operating the biometric authentication device are secured. The entire apparatus control program 207 is a program for controlling the entire biometric authentication apparatus 101. The peripheral device I / O control program 208 controls the peripheral device I / O device 202.

  The authentication program 209 performs feature extraction processing on the biometric image output from the image sensor 204, and compares the extracted feature value with a template that is a reference feature value for comparison stored in the biometric authentication device. . Then, as a result of the collation, collation acceptance or collation rejection is output, and a collation distance value representing the difference between the extracted feature quantity and the template is output.

  The collation distance value conversion program 210 inputs the collation distance value output by the authentication program 209 and performs a process of converting it into a collation distance value for output to a computer. Specifically, a random number or the like is added to the collation distance value output by the authentication program 209, and the collation distance value is processed for output. Instead of the collation distance value, an expression “confidence value” indicating the degree of conviction that the person is identifiable may be used.

  The display image generation program 211 is processed so that it is easy for a user of biometric authentication to read, such as inputting a biometric image or a biometric feature that the authentication program 209 has determined acceptance / rejection of authentication and cutting out only the biometric part. To generate a biological image for display.

  Thus, the program has various functions, performs various processes, and is controlled by the hardware configuration of the control unit 201 as described above. In the present invention, the description will focus on programs, but various functions of these programs, for example, control means, authentication means, collation distance value conversion means, display image generation means, etc. can be expressed as each part. Needless to say.

  The data storage unit 212 includes (1) biometric image data (raw data) acquired by the image sensor 204, (2) biometric feature amount generated by extracting only a vein pattern from the biometric image data by the authentication program 209, (3 ) A template that is a biometric feature serving as a reference for the person for comparison and collation with the biometric feature, (4) a collation distance value threshold that is a threshold for the collation distance value for determining authentication acceptance / rejection, ( 5) As a result of comparing and collating the biometric feature quantity with the template, a collation distance value indicating the degree of difference between them (if the collation distance value is smaller than the threshold value, it is determined that the authentication is accepted, and if the collation distance value is greater than the threshold value, the collation rejection is (6) Matching result indicating the result of the authentication program 209 determining whether to accept / reject authentication (stores “1” for acceptance, “0” for rejection), ( The collation distance value conversion program 210 inputs the collation distance value, and the output collation distance value subjected to the conversion process (sent to the computer 102 and used for screen display), (8) the display image generation program 211 Is an area for storing a biological image for display on the computer 102 to which the biological image stored in the data storage unit 212 is input.

  FIG. 3 is a diagram illustrating a configuration example of the computer 102. A control unit (CPU) 301 is a processor responsible for terminal data processing and various controls. The peripheral device I / O device 302 is an interface for connecting to the biometric authentication device 101. A display device 303 is a monitor that displays a biometric authentication result, a biometric image and a biometric feature used for authentication, and a key input device 304 is an operator key input device. Various programs and data are stored in the main storage device 306. A bus 305 connects each device in the terminal.

  The main storage device 306 controls the overall authentication program 307 for controlling the computer 102, the biometric authentication device 101 connected via the peripheral device I / O device 302, processes a biometric image for display, A biometric authentication application program 308 for displaying authentication results and images on the display device 303, a peripheral device I / O control program 309 for controlling the peripheral device I / O device 302, and a process suitable for the biometric image stored in the data storage unit 311 And a display image processing program 310 for outputting a processed biological image is stored.

  The data storage unit 311 includes (1) a display biometric image in which data of a display biometric image stored in the biometric authentication device 101 is transferred via the peripheral device I / O device 302, and (2) a display image processing program. The output processed biometric image for display, (3) the collation result stored in the biometric authentication device 101 of FIG. 2, and (4) the collation distance value for output stored in the biometric authentication device 101 of FIG. Remember.

  FIG. 4 is a flowchart of operations of the biometric authentication device 101 and the computer 102.

  Based on an operation from a staff member, the control unit 301 of the computer 102 transmits an activation signal by using the stored biometric authentication application program 308 to activate the biometric authentication device 101 and instruct to execute a matching process ( Step 401).

  When the control unit 201 of the biometric authentication apparatus receives an execution instruction of the verification process from the computer 102, the control unit 201 controls the illumination LED 203 and the image sensor 204 to acquire a user's biometric image and store it in the data storage unit 212. (Step 402). Thereafter, the control unit 201 of the biometric authentication device reads the biometric image stored in the data storage unit 212, extracts the biometric feature amount from the image, and stores it in the data storage unit 212 (step 403).

  Next, the control unit 201 of the biometric authentication device reads out the biometric feature amount stored in the data storage unit 212 in step 403 and the template stored in advance in the data storage unit 212 and obtains a matching distance value between them. Calculate (matching and collation processing) and store in the data storage unit 212. Further, the collation distance value is compared with a preset threshold value, and if it is less than the threshold value, the collation is accepted, and if it is greater than or equal to the threshold value, the collation rejection value is stored in the data storage unit 212 (step 404). Note that these steps 402 to 404 are processes executed using the authentication program 209.

  Thereafter, the control unit 201 of the biometric authentication apparatus reads the collation distance value from the data storage unit 212, adds a random number or the like to convert it, generates a converted collation distance value (also referred to as an output collation distance value), and stores the data. The data is stored in the unit 212 (step 405). Here, in the case of collation acceptance, a random number that is less than the threshold value is added to the collation distance value so that the collation result and the collation distance value do not contradict each other. The same applies to the case of collation rejection, and a random number that exceeds the threshold is added.

  In the case of collation rejection, a random number that is equal to or greater than the collation distance value threshold is added. FIG. 5 shows an example of specific processing. It is assumed that the collation distance value stored in the data storage unit 212 is 0.19 and the threshold value is 0.09. Since the difference 501 between the collation distance value and the threshold value is 0.10, a uniform random number having a random number width 502 centered on 0.19 and 0.20 is generated and used as the collation distance value for output (FIG. 5 ( a)).

  On the other hand, in the case of collation acceptance, a random number such that the collation distance value is less than the threshold value 0.09 is added to the collation distance value so that the collation result and the collation distance value do not contradict each other. In FIG. 5B, since the difference 503 from the threshold value is 0.07 when the collation distance value is 0.02, a uniform random number with a width of 0.14 centered on 0.02 is generated, Use the output collation distance value. However, since the collation distance value may be a negative value, a uniform random number between 0.0 and 0.09 is generated as indicated by 504, and this is the collation distance for output (after conversion). Value. By performing these processes, a transition of the collation distance value as shown in FIG. 6 is obtained, and even if a hill climbing attack using the collation distance value is attempted, the collation distance value changes randomly. There are no clues and the attack becomes extremely difficult.

  In the above description, a uniform random number having a width twice as large as the difference between the threshold value and the collation distance value is generated around the collation distance value, and used as the collation distance value after conversion. However, the width of the uniform random number decreases as the collation distance value approaches the threshold value, and the difference between the collation distance value before conversion and the collation distance value after conversion decreases near the threshold value, so that the original value can be easily estimated. Therefore, a lower limit such as 0.10 is set for the width in which uniform random numbers are distributed, and in the case of collation rejection, uniform random numbers between the threshold value and the threshold value +0.10 may be generated and used as the collation distance value after conversion. . In this case, the collation distance value before conversion is no longer the center of the uniform random number. The conversion method at the time of collation acceptance is the same, and a uniform random number between the threshold value and the threshold value −0.10 may be generated and used as the collation distance value after conversion.

  The collation distance value may be converted using a random number only when the biometric detection function detects that the artificial object is held over the biometric authentication device. Here, the biometric detection function refers to a biometric detection sensor provided in the biometric authentication device or acquisition of whether an image acquired by the biometric authentication device is derived from a living body or an artifact. It is a function to detect using the image information. As a living body detection method, several known methods such as temperature, electric resistance, capacitance, and the like are known. In order to develop a counterfeit organism, it is expected that the artifact is held over the biometric authentication device many times. In this case, it is desired to convert the collation distance value and make the artifact development difficult. On the other hand, when holding the living body, since it is the original correct operation, it is easier to operate if the collation distance value is output as it is. Therefore, it is desirable to control the conversion of the collation distance value by the output of the living body detection function.

  Further, the control unit 201 of the biometric authentication device reads the biometric image used for the verification acceptance / rejection determination from the data storage unit 212 and transmits it to the computer 102 together with the verification result and the converted verification distance value (step 406). The display image generation program 211 may be executed to perform appropriate image processing that is easy for the biometric user to view, such as deleting a background portion in the image.

  On the other hand, the control unit 301 of the computer receives the biometric image, the collation result, and the converted collation distance value and stores them in the data storage unit 311 (step 407). In this way, the collation process is completed by a series of processes in steps 402 to 406.

  The control unit 301 of the computer executes the display image processing program 310 to perform an appropriate image processing for preventing a hill climbing attack on the biological image stored in the data storage unit 311, and the processed biological image is obtained. The data is stored in the data storage unit 311 (step 408). Finally, the collation result, the collation distance value for output, and the processed biological image are displayed on the display device 303 (step 409).

  FIG. 7 shows an example of image processing in step 408. FIG. 7A shows a biometric image before processing. A black portion around the image is a background, a white portion is a part of a finger, and a thin curve or a thick curve in a white portion represents a vein pattern. One or more appropriate portions are randomly selected from the vein pattern, and the blood vessel pattern of the portion is thickened or thinned.

  In FIG. 7B, one or more appropriate portions are randomly selected in the vein pattern, and the blood vessel pattern of the portion is thickened or thinned. A rectangle surrounded by a dotted line is an image conversion region selected at random, and the vein pattern in the rectangle is thickened. As a result, an image difference in the rectangular portion always occurs between the image before processing in FIG. 7A and the image after processing in FIG. 7B, and the position where the image difference appears is different each time authentication is performed. As a result, even if an attempt is made to improve a counterfeited living body so that the same image as the target living body can be obtained using the display image of the biometric authentication device, the direction of improvement cannot be understood.

  Here, in order to generate a random image difference, the direction in which the blood vessel passes is maintained only by thickening or narrowing the blood vessel at an arbitrary position, so that visibility during viewing is not impaired. On the other hand, FIG. 7C is a process that generates an image difference from FIG. 7A, but the rectangular blood vessel direction surrounded by the dotted line that is the process target is different from FIG. 7A. The user tends to feel uncomfortable when visually checking.

  FIG. 8 shows another example of image processing for maintaining the blood vessel direction. FIG. 8A shows an example of processing by changing the image luminance value of the blood vessel background portion locally as in 801 to 805. FIG. 8B shows an example in which the gradation of the blood vessel background portion is changed. In either case, the blood vessel traveling direction is maintained, so that visibility is hardly impaired. It should be noted that any image processing method for maintaining such a blood vessel direction may be used as long as the arbitrarily defined collation distance value changes in order to prevent the hill climbing attack described above. .

  FIG. 9 is a screen display example of the authentication result in step 409. The collation result, the collation distance value for output, and the processed biometric image for display are displayed on the display device 303. FIG. 9A shows the result when collation is rejected, and FIG. 9A shows a screen example when accepting collation. Collation distance values 0.12 and 0.07 indicate collation distance values converted using random numbers as shown in FIG. A portion surrounded by a dotted line in the display images 901 and 902 indicates that image processing has been performed, and the dotted line is not displayed on an actual screen.

  Finally, problems in the prior art and the present embodiment corresponding thereto will be described. In the prior art, a counterfeit organism could be developed by a hill climbing attack. For a hill climbing attack, some numerical index that is a guideline for improving a counterfeit organism, such as a collation distance value, is necessary.

  FIG. 10A shows an example of a change in collation distance value associated with a hill climbing attack. As feedback information, the case where the collation distance value is abused for the development of counterfeit organisms is assumed. First, when an arbitrary counterfeit biological body is held over the biometric authentication device, the biometric authentication device acquires an image, extracts a feature amount, compares it with a registered template, and calculates a collation distance value. As a matter of course, since the collation distance value exceeds the threshold value, collation is rejected. Next, an arbitrary improvement is applied to the forged biometric, and the verification distance value is calculated by holding it over the biometric authentication device again in the same manner as described above. As a result, if the value is smaller than the previous collation distance value, the improvement is considered correct, and the same improvement is continued again. To improve. This series of counterfeit biological improvement cycles is repeated many times until the collation distance value falls below the threshold value.

  On the other hand, as one method for preventing such a hill climbing attack, there is a method of outputting only a discrete value instead of a continuous value as a collation distance value. FIG. 10B shows an example in which the change in the collation distance value due to the hill climbing attack is converted into a discrete value.

  FIG. 11 shows a table for converting the collation distance value d, which is a continuous value, into a discrete value. For example, the second row from the top indicates that all collation distance values satisfying the condition of 0.025 ≦ d <0.075 are converted to a discrete value of 0.05.

  Even if the counterfeit organism is improved and the collation distance value is slightly reduced, if only a discrete value can be obtained, the collation distance value does not change. For this reason, it is impossible to determine whether the forged body is improved or not, and development efficiency is deteriorated.

  However, even such a method of taking a discrete value is not necessarily a universal one because a hint for improvement can be given where a change in the discrete matching distance value occurs. Furthermore, since a third party having malicious intent defines the collation distance value, a discrete value cannot be taken in the first place.

  Even when the collation distance value cannot be obtained, such a numerical index can be generated from the image displayed on the screen.

  For example, when registering a template or when authentication succeeds, biometric information (sensor output image) displayed on the screen, biometric feature image using a function such as “Print Screen” of Windows (registered trademark), Save as image A. Next, the counterfeit body is held over the sensor, and the biological information (image) and biological feature amount (image) are similarly stored as an image B using a function such as “Print Screen” of Windows (registered trademark). Furthermore, a collation distance value representing an image difference between the image A and the image B is defined in some form, for example, an absolute value difference of pixel values between the image A and the image B is accumulated for each pixel, and the value is It is defined as a collation distance value. Then, the counterfeit organism is gradually improved in the same manner as described above so that the collation distance value becomes small. If the counterfeit organism is improved many times while looking at the collation distance value, the collation distance value gradually decreases and the collation distance value similarly decreases. Finally, if the collation distance value is smaller than the threshold value for determining the identity, the forged body can be completed.

  On the other hand, if the configuration of the present embodiment is adopted, it becomes extremely difficult to develop a counterfeit organism using feedback information displayed on the display device 303. That is, even if you try to improve the counterfeit organism little by little to make the verification distance value smaller, the system will change the verification distance value arbitrarily and randomly, so that the change in the verification distance value is due to the improvement of the counterfeit organism It is unclear whether the collation distance value is arbitrarily changed on the system side. As a result, the direction of improvement of counterfeit organisms is not understood, and development of counterfeit organisms becomes extremely difficult.

  In addition, even if an attempt is made to improve a counterfeit organism using the same displayed biological image or biological feature amount, the displayed biological image and feature amount can be arbitrarily changed. Accordingly, the collation distance value obtained from the biometric image or the biometric feature amount is arbitrarily and randomly changed. As a result, it is not clear whether the change in the collation distance value is due to the improvement of the counterfeit organism or the display image is arbitrarily changed on the system side. Therefore, in the same way as described above, the direction of improvement of counterfeit organisms is not known, and the effect of preventing forgery organism development can be obtained.

  In the first embodiment, the generation of the output collation distance value is performed by the biometric authentication device 101. However, these processes may be performed by the computer 102. In that case, it is necessary to store the collation distance value conversion program 210 and the display image generation program 211 in the main storage device 306 of the computer 102 and to transmit the collation distance value and the image data in the data storage unit 212 to the computer 102. That is, the process of step 405 in FIG.

  In the first embodiment, the generation of the display biometric image is performed by the computer 102. However, these processes may be performed on the biometric authentication apparatus 101 side. In that case, it is necessary to store the display processing program 310 in the main storage device 206 of the biometric authentication device 101. That is, the process of step 408 in FIG.

  DESCRIPTION OF SYMBOLS 101 ... Biometric authentication apparatus, 102 ... Computer, 201 ... Control part (CPU), 202 ... Peripheral device I / O device, 203 ... Illumination LED, 204 ... Image sensor, 205 ... Bus, 206 ... Main memory, 207 ... Device Overall control program, 208 ... Peripheral device I / O control program, 209 ... Authentication program, 210 ... Verification distance value conversion program, 211 ... Display image generation program, 212 ... Data storage unit, 301 ... Control unit (CPU), 302 ... peripheral device I / O device, 303 ... display device, 304 ... key input device, 305 ... bus, 306 ... main storage device, 307 ... overall control program, 308 ... biometric authentication application program, 309 ... peripheral device I / O control Program, 310 ... Display image processing program, 311 ... Data storage unit

Claims (8)

A biometric authentication device that performs personal identification using personal biometric information,
Biometric information reading means for reading biometric information;
Feature quantity extraction means for extracting biometric feature quantities from the biometric information read by the biometric information reading means;
Storage means for storing biometric features that are reference data for authentication;
Collation that collates the biometric feature amount extracted by the feature amount extraction unit with the biometric feature amount stored in advance in the storage unit and outputs a collation distance value indicating the degree of coincidence between the two biometric feature amounts Means,
Collation distance value conversion for converting the collation distance value output from the collation means into an arbitrary value within the range smaller than the threshold when the collation is accepted, and to an arbitrary value within the range equal to or greater than the threshold when the collation is rejected Means,
A biometric authentication device comprising: a transmission unit that transmits the verification distance value converted by the verification distance value conversion unit to a computer that controls the biometric authentication device. The biometric authentication device according to claim 1,
The biometric authentication device, wherein the collation distance value converting means converts the collation distance value into a predetermined value using a random number. The biometric authentication device according to claim 2, wherein
The biometric authentication device, wherein the collation distance value conversion means provides a lower limit value for the random number when the collation distance value is converted using a random number. A biometric authentication system configured by a biometric authentication device that performs identity verification and a computer that controls the biometric authentication device,
The biometric authentication device is:
Biometric information reading means for reading biometric information;
Feature quantity extraction means for extracting biometric feature quantities from the biometric information read by the biometric information reading means;
Storage means for storing biometric features that are reference data for authentication;
Collation that collates the biometric feature amount extracted by the feature amount extraction unit with the biometric feature amount stored in advance in the storage unit and outputs a collation distance value indicating the degree of coincidence between the two biometric feature amounts Means,
Collation distance value conversion for converting the collation distance value output from the collation means into an arbitrary value within the range smaller than the threshold when the collation is accepted, and to an arbitrary value within the range equal to or greater than the threshold when the collation is rejected Means,
Transmission means for transmitting the collation distance value converted by the collation distance value conversion means to a computer that controls the biometric authentication device,
The computer
A biometric authentication system comprising storage means for receiving and storing the collation distance value transmitted from the transmission means of the biometric authentication device. The biometric authentication system according to claim 4, wherein
The biometric authentication system, wherein the computer further includes display means for displaying a collation distance value stored in the storage means. A biometric authentication system configured by a biometric authentication device that performs identity verification and a computer that controls the biometric authentication device,
The biometric authentication device is:
Biometric information reading means for reading biometric information;
Feature quantity extraction means for extracting biometric feature quantities from the biometric information read by the biometric information reading means;
Transmission means for transmitting the collation distance value extracted by the feature amount extraction means to the computer,
The computer
Storage means for storing biometric features that are reference data for authentication;
Collation means for collating the biometric feature quantity stored in advance in the storage means with the biometric feature quantity received from the biometric authentication device, and calculating a collation distance value indicating the degree of coincidence of the two biometric feature quantities; ,
Collation distance value conversion for converting the collation distance value calculated by the collation means into an arbitrary value within the range smaller than the threshold when the collation is accepted, and to an arbitrary value within the range above the threshold when the collation is rejected Means,
A biometric authentication system comprising: a display unit that displays the collation distance value converted by the collation distance value conversion unit. The biometric authentication system according to any one of claims 4 to 6,
The verification distance value converting means converts the verification distance value into a predetermined value using a random number. The biometric authentication system according to claim 7, wherein
The biometric authentication system, wherein the collation distance value conversion means provides a lower limit for the random number when the collation distance value is converted using a random number.

Images