JP2008009617A - System, program, and method for individual biological information collation - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To highly and suitably execute collation using individual biological information. <P>SOLUTION: In an individual biological information collation system in which data related to an individual in a database storing the data related to the individual and individual biological information related to the designated person are collated and the collated results are outputted, the database storing a feature quantity of individual biological information is held as the data related to the individual, a feature quantity of individual biological information is calculated from the individual biological information related to the designated individual, a collation score indicating the matching degree among individual biological information is calculated by comparing the feature quantity of the calculated individual biological information with the feature quantity of the individual biological information stored in the database, and the collated results determined by the calculated collation score is outputted to a designated output section. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a personal biometric information collation apparatus, a personal biometric information collation program, and a personal biometric information collation method.

  Conventionally, in the situation of a financial consumer loan agreement between a general consumer (hereinafter referred to as a customer) and a financial institution, the financial institution overlaps with the customer before completion of repayment, which has not yet been paid back. Therefore, there is an operation form in which the contract is not concluded. For this reason, a customer who has been impersonated by using a forged driver's license, despite being a customer before the completion of repayment, for the purpose of avoiding such restrictions due to the operation mode, There is a situation in which the contracts are concluded repeatedly.

  As a technique for preventing the occurrence of such a situation, a technique for confirming whether or not a customer who wishes to conclude a contract is a customer before completion of repayment by using customer face image data is conceivable. For example, the technique disclosed in Patent Document 1 is a technique that uses part image data obtained by dividing face image data. Specifically, first, part image data (for example, image data of eyes, nose, mouth, contour, etc.) obtained by dividing the face image data of a customer who wishes to conclude a contract is input to the device, and then the device is The inputted part image data is collated with the part image data of the customer before completion of repayment stored in advance in the database. Then, the apparatus determines whether or not the customer of the input part image data may be the same person as the customer before the repayment is completed, and may be the same person. The determined face image data of the customer before completion of the repayment is output as a matching result.

JP 2003-331264 A

  By the way, as described below, the above-described conventional technique has a problem in that collation using personal biometric information cannot be executed in a sophisticated and appropriate manner. That is, for example, when part image data of “eyes” are collated, even if “whole face” is not similar, it is determined that there is a possibility that face image data having only similar “eyes” may be the same person. Therefore, there is a possibility that the result of the collation may be satisfied, and there is a possibility that the range of the collation result may be inappropriately widened. On the other hand, if the criteria are tightened to prevent the situation where the range of verification results spreads improperly, there is a risk that the verification results will deviate from the results of the verification only due to differences over time or the presence or absence of glasses. There is a risk of narrowing. Therefore, in the method using the part image data obtained by dividing the face image data, the matching using the personal biometric information cannot be executed at a high level and appropriately.

  The above-mentioned issues are not limited to the case where a customer impersonating another person concludes a duplicate contract with a financial institution, for example, a scene of illegal entry by a person who is impersonated as a separate person using a forged passport, The same problem arises in situations where it is necessary to check whether or not there is a possibility that the person is the same person, such as a test-take-out test by a student who impersonates another person using a fake admission ticket.

  Accordingly, the present invention has been made to solve the above-described problems of the prior art, and a personal biometric information collation apparatus and personal biometric information capable of executing collation using personal biometric information in a sophisticated and appropriate manner. An object is to provide a collation program and a personal biometric information collation method.

  In order to solve the above-described problems and achieve the object, the invention according to claim 1 is directed to a database for storing data relating to an individual and storing the data related to the individual, and an individual living body related to a predetermined individual. A personal biometric information collating apparatus that collates information and outputs a result of the collation, wherein the database holding means holds the database that stores the feature quantity of the personal biometric information of the individual as data related to the individual; Feature quantity calculation means for calculating the feature quantity of the personal biometric information from the personal biometric information related to the predetermined individual, and the feature quantity of the personal biometric information calculated by the feature quantity calculation means and held by the database holding means The collation score for calculating the collation score indicating the degree of collation between the personal biometric information by comparing with the feature quantity of the personal biometric information stored in the database. And A calculating means, characterized in that and an output means for outputting the results of the verification on a predetermined output unit are determined by the matching score calculated by the matching score calculating means.

  Further, in the invention according to claim 2, in the above invention, the database holding unit holds the database that stores a feature amount of a face image obtained by capturing a face as the feature amount of the personal biometric information, and the feature The amount calculating unit calculates a feature amount of the face image as the feature amount of the personal biometric information, and the matching score calculating unit stores the feature amount of the face image calculated by the feature amount calculating unit and the database. The comparison score is calculated by comparing with the feature amount of the face image.

  The invention according to claim 3 further includes an identification image capturing means for capturing an identity certificate issued by a public institution in which personal biometric information is described in the above invention, wherein the feature amount calculating means includes: A feature amount of the personal biometric information is calculated from the personal biometric information captured by the identification image capturing means.

  The invention according to claim 4 is the above-mentioned invention, further comprising personal imaging means for imaging personal biometric information of the individual who is the subject, wherein the feature amount calculating means is the individual imaged by the personal imaging means. The feature amount of the personal biometric information is calculated from the biometric information.

  According to a fifth aspect of the present invention, in the above invention, the database held by the database holding means includes the personal biometric information in which the feature quantity of the personal biometric information is described in the captured identification card. And the personal identification information calculated based on the personal biometric information calculated from the personal biometric information obtained by capturing the personal biometric information of the individual who is the subject, and storing the collation score When the feature amount of the personal biometric information is calculated by the feature amount calculation unit from the personal biometric information captured by the identification image capturing unit, the calculation unit calculates the feature amount and the identification-based feature amount Are compared with each other to calculate a matching score, and the feature quantity of the personal biometric information is calculated by the feature quantity calculation means from the personal biometric information imaged by the personal imaging means. Case, and calculates a matching score by comparing the said characteristic quantity and the individual system feature amount.

  According to a sixth aspect of the present invention, in the above invention, the feature quantity of the personal biometric information calculated by the feature quantity calculation means from the personal biometric information imaged by the identification image pickup means and the personal imaging means Comparing the feature quantity of the personal biometric information calculated by the feature quantity calculating means from the personal biometric information imaged by the personal biometric information, and calculating an identification check collation score indicating the degree of collation between the personal biometric information A score calculation means is further provided.

  Further, in the invention according to claim 7, in the above invention, the output means may provide, as a result of the collation, an individual compared with the collation score calculated by the collation score calculation means when calculating the collation score. The data is output to the predetermined output unit in association with the data.

  In the invention according to claim 8, in the above invention, the output unit arranges the results of the collation in descending order of the collation score values calculated by the collation score calculation unit, and outputs the collation results to the predetermined output unit. It is characterized by outputting.

  The invention according to claim 9 is the above invention, wherein the output means is a gauge, meter, mark, color, font, which visually expresses the value of the collation score calculated by the collation score calculation means, The verification result is output to the predetermined output unit in a form processed using any one or a plurality of sizes.

  Further, in the invention according to claim 10, in the above invention, the output means, when the identification confirmation collation score calculated by the identification confirmation collation score calculation means exceeds a predetermined threshold, A predetermined display indicating that the value is exceeded is output to the predetermined output unit.

  According to an eleventh aspect of the present invention, in the above invention, the personal imaging unit continuously captures the individual's face image until a predetermined time elapses after the start of imaging, and the identity verification check The score calculating means calculates the identity confirmation collation score using the feature amount of the front face image when the personal imaging means captures the front face image of the individual, and the output means When a plurality of front face images are imaged by the personal imaging means, and a plurality of identification confirmation collation scores are calculated according to the plurality of front face images by the identification confirmation collation score calculation means, the plurality of identity confirmations A combination of the face image of the front of the individual compared in calculating the highest value identification confirmation matching score and the highest value identification confirmation matching score among the matching scores, One or more of a combination of the face image of the front of the person compared in calculating the lowest value identification check score and the lowest value confirmation check score in the minute check reference score is the predetermined value. It outputs to an output part, It is characterized by the above-mentioned.

  The invention according to claim 12 is the above invention, wherein the personal imaging means continuously captures the personal face image until a predetermined time has elapsed since the start of imaging, and the identification check The score calculation means captures the plurality of personal identification verification scores when the personal imaging means captures a plurality of individual facial images and calculates a plurality of identification verification verification scores according to the plurality of facial images. An average value is calculated, and the output means outputs the average value to the predetermined output unit when the average value is calculated by the identity confirmation collation score calculation means.

  The invention according to claim 13 is the above invention, wherein personal biometric information captured by the identification image capturing means, feature amounts calculated by the feature amount calculating means from the personal biometric information, and the personal image capturing means Personal biometric information captured by the personal biometric information, a feature quantity calculated by the feature quantity calculation means from the personal biometric information, an identity confirmation collation score calculated by comparing the two feature quantities by the identity confirmation collation score calculation means, It further comprises registration means for associating and registering any one or more of these in the database.

  Further, the invention according to claim 14 collates the data related to the individual in the database storing the data related to the individual for each individual and the personal biometric information related to the predetermined individual, and the result of the verification is obtained. A personal biometric information collation program for causing a computer to execute an output method, the database holding procedure for holding the database for storing the feature quantity of the personal biometric information of the individual as data relating to the individual, and for the predetermined individual The feature amount calculating procedure for calculating the feature amount of the personal biometric information from the personal biometric information, the feature amount of the personal biometric information calculated by the feature amount calculating procedure, and the database held by the database holding procedure are stored. Collation that compares the feature quantity of personal biometric information and calculates a collation score indicating the degree of collation between personal biometric information A core calculation procedure, characterized in that to execute an output procedure for outputting the result of the verification to the predetermined output unit which are determined by the matching score calculated, to a computer by the matching score calculation procedure.

  Further, the invention according to claim 15 collates the data related to the individual in the database storing the data related to the individual for each individual and the personal biometric information related to the predetermined individual, and the result of the verification is obtained. A personal biometric information collation method for outputting, comprising: a database holding step of holding the database for storing feature quantities of personal biometric information of the individual as data relating to the individual; and the personal biometric information relating to the predetermined individual A feature amount calculating step for calculating a feature amount of personal biometric information, a feature amount of the personal biometric information calculated by the feature amount calculating step, and a feature amount of the personal biometric information stored in the database held by the database holding step And a collation score calculation step of calculating a collation score indicating the degree of collation between the personal biometric information, and the collation score calculation An output step of outputting the result of said verification as determined by said calculated matching score to a predetermined output unit by extent, characterized in that it contains.

  According to the invention of claim 1, claim 14, or claim 15, data relating to individuals in a database that stores data relating to individuals divided by individual and personal biometric information relating to a predetermined individual are collated And a personal biometric information collation device for outputting a collation result, storing a database for storing a feature amount of the personal biometric information of the individual as data relating to the individual, and from the personal biometric information relating to the predetermined individual The feature amount of the personal biometric information is calculated, and the feature amount of the calculated personal biometric information is compared with the feature amount of the personal biometric information stored in the stored database to calculate a matching score indicating the degree of matching between the personal biometric information. Since the result of the collation determined by the calculated collation score is output to a predetermined output unit, for example, personal biometric information of the customer who wishes to conclude a contract (for example, the entire face image of the customer) And the feature amount calculated from the personal biometric information of the customer before completion of repayment stored in the database in advance, the customer who wishes to conclude a contract with the customer before completion of repayment Compared with the method that uses part image data obtained by dividing face image data, the verification using personal biometric information is advanced and appropriate because it is a method that checks whether or not there is a possibility of being the same person. It becomes possible to execute.

  According to the invention of claim 2, the database storing the feature quantity of the face image obtained by capturing the face is stored as the feature quantity of the personal biometric information, and the feature quantity of the face image is used as the feature quantity of the personal biometric information. Since the comparison score is calculated by comparing the calculated feature value of the face image with the feature value of the face image stored in the database, the method of matching based on the feature value of the face image obtained by capturing the face is: Compared to the method of collating based on the feature quantity of other personal biometric information such as fingerprint feature quantity and vein feature quantity, it is easier to collect the predetermined data used for calculating the feature quantity. Collation using information can be executed in an advanced, appropriate and simple manner.

  According to the third aspect of the present invention, since the identity certificate issued by a public institution in which the personal biometric information is described is imaged, and the feature quantity of the personal biometric information is calculated from the captured personal biometric information. Since the face image used for calculating the feature amount of the face image is obtained by capturing an ID issued by a public institution (eg, driver's license, passport) with a scanner, etc. Collation using information can be executed in an advanced, appropriate and simple manner.

  According to the invention of claim 4, since the personal biometric information of the individual who is the subject is imaged and the feature amount of the personal biometric information is calculated from the captured personal biometric information, for example, the feature amount of the face image is calculated Since the face image used in the above is acquired by capturing an individual's face as a subject with a camera or the like, collation using personal biometric information can be executed in an advanced, appropriate and simple manner.

  According to the invention of claim 5, the retained database includes the feature quantity of the personal biometric information, and the identity-certificate-type feature quantity calculated from the personal biometric information described in the captured identification card. The personal biometric information of the individual who is the subject is classified and stored in the personal system feature amount calculated from the captured personal biometric information, and from the personal biometric information described in the captured identification card When the feature amount of the personal biometric information is calculated, the feature score is compared with the feature value of the identification card system to calculate the matching score, and the feature amount of the personal biometric information is calculated from the captured personal biometric information of the individual In this case, since the comparison score is calculated by comparing the feature quantity with the personal feature quantity, for example, the feature quantities calculated from the face photograph of the driver's license taken with a scanner or the like, or the camera, etc. Individuals imaged at The method of calculating the matching score by comparing the feature quantities calculated from the facial photos reduces the difference in the environment such as the imaging conditions, for example, calculated from the facial photo of the driver's license taken with a scanner etc. Compared with the technique of calculating the matching score by comparing the feature quantity with the feature quantity calculated from the personal face photograph taken with a camera etc., matching using personal biometric information is advanced, appropriate and accurate It becomes possible to execute.

  According to the invention of claim 6, the feature amount of the personal biometric information calculated from the personal biometric information described in the captured identity card and the individual calculated from the personal biometric information of the captured individual Before comparing the data related to individuals in the database with the personal biometric information related to a predetermined individual, since the identification check score indicating the degree of matching between the personal biometric information is calculated by comparing with the feature amount of the biometric information. As a stage, for example, a face photo of a driver's license taken with a scanner or the like and a personal face photo taken with a camera or the like are collated to confirm the identity (identity confirmation), and a predetermined individual to be collated Since it is confirmed that the personal biometric information related to the personal data is the person's data, it is possible to execute collation using the personal biometric information at a high level and more appropriately.

  According to the invention of claim 7, as a result of the collation, the calculated collation score and the data relating to the individual compared in the calculation of the collation score are output to the predetermined output unit in association with each other. A collation score and a face image are output in correspondence with a predetermined output unit (monitor), and the result of the collation is presented to an operator or the like, so that collation using personal biometric information is advanced, appropriate and suitable for operation. It becomes possible to execute with.

  According to the invention of claim 8, since the collation results are arranged and output to a predetermined output unit in descending order of the calculated collation score value, for example, the collation score is output to a predetermined output unit (monitor). Face images are arranged and output in descending order, and verification is performed from the highest collation score value by an operator or the like, so that collation using personal biometric information can be executed at an advanced, appropriate and efficient level. Become.

  According to the invention of claim 9, in a form processed using any one or more of a gauge, a meter, a mark, a color, a font, and a size for visually expressing the calculated collation score value, Since the result of the matching is output to a predetermined output unit, for example, the value of the matching score can be increased by, for example, presenting the value of the matching score with a gauge or presenting a face image with a high value of the matching score. Since it is visually expressed and the possibility that an operator or the like may leak is reduced, collation using personal biometric information can be executed with high accuracy, appropriateness, and reliability.

  According to the invention of claim 10, when the calculated identity confirmation collation score exceeds a predetermined threshold value, a predetermined display indicating that the predetermined threshold value is exceeded is output to a predetermined output unit. For example, since the result of determining whether or not the user is a person according to a predetermined threshold is output, and the burden of confirmation by the operator is reduced, collation using personal biometric information is executed at an advanced and appropriate level with a reduced operational burden. It becomes possible to do.

  According to the invention of claim 11, when a face image of an individual is continuously captured until a predetermined time has elapsed since the start of imaging, and a face image of the front of the individual is captured, When an identification check collation score is calculated using the feature amount of the face image, a plurality of front face images are captured, and a plurality of identification confirmation collation scores are calculated according to the plurality of front face images, The combination of the face image of the front of the person compared when calculating the highest ID verification score and the highest ID verification score, among the multiple ID verification scores Since one or more of the combinations of the face images of the front faces of the individuals compared at the time of calculating the value identification verification score and the lowest value identification verification score are output to a predetermined output unit, for example, When capturing a person's face, which is a body, with a camera or the like, multiple photographs are taken consecutively to prevent imaging errors and the front face necessary for matching is detected. It is possible to execute the used verification in a sophisticated, appropriate and reliable manner. In addition, when a large number of face images are captured and identity verification (identity confirmation) is performed, only a small amount of important verification results are extracted from the verification results and used for verification by the operator. Collation using information can be executed in an advanced, appropriate and efficient manner. Furthermore, for example, if the operator confirms only the best photograph (the face image with the highest matching score), the possibility of mistakenly identifying another person who is similar is the worst photograph. If the operator confirms only (the face image with the lowest matching score), there is a higher possibility that a dissimilar person will be mistakenly confirmed not to be the person, but which policy will be used for confirmation? Because it is output according to the policy according to the usage pattern, such as whether to check with a policy that combines both, or it is possible to perform collation using personal biometric information in a sophisticated, appropriate and operational format Become. In addition, since it is possible to quantitatively determine the appearance of the face image using the matching score with the face photo of an ID card such as a driver's license, it is possible to execute matching using personal biometric information in a sophisticated and appropriate manner. It becomes possible.

  According to the twelfth aspect of the present invention, an individual's face image is continuously captured until a predetermined time has elapsed since the start of imaging, and a plurality of the individual face images are captured, and a plurality of face images are captured. When a plurality of identification confirmation matching scores are calculated according to the above, the average value of the plurality of identification confirmation matching scores is calculated, and when the average value is calculated, the average value is output to a predetermined output unit. Since an average collation result (average value) is presented to the operator from a large amount of collation results, collation using personal biometric information can be executed in a sophisticated, appropriate and efficient manner.

  According to the invention of claim 13, the personal biometric information of the captured identification card, the feature amount calculated from the personal biometric information, the personal biometric information of the captured individual, and the feature amount calculated from the personal biometric information Since one or more of the identification confirmation collation scores calculated by comparing the two feature quantities are registered in the database, for example, the face when the identity confirmation (identification confirmation) is performed by the operator When the image is registered in the database, it can be reconfirmed later by another person, so that collation using personal biometric information can be executed in an advanced, appropriate and reliable manner.

  Exemplary embodiments of a personal biometric information collation apparatus, a personal biometric information collation program, and a personal biometric information collation method according to the present invention will be described below in detail with reference to the accompanying drawings. The main terms used in the following embodiments, the outline and features of the personal biometric information collating apparatus according to the first embodiment, the configuration and processing procedure of the personal biometric information collating apparatus according to the first embodiment, and the effects of the first embodiment. This will be described in order, and then another embodiment will be described.

[Explanation of terms]
First, main terms used in the following examples will be described. The “database” is usually constructed by storing a large amount of data on a large-capacity hard disk. The “database” is constructed according to the purpose of use of the database, such as what kind of data is stored or how the data is divided and stored. In the following embodiments, the “database” is constructed for the purpose of managing information related to an individual, and therefore stores data related to an individual classified by individual.

  Here, “individual data” refers to specific values corresponding to a specific individual regarding attributes related to the individual such as “ID”, “name”, and “address”. That is, for example, “1”, “Tokyo Taro”, “Tokyo ...”, and the like. In addition, “divide and store personal data for each individual” means that the database corresponds to a specific individual such as “1”, “Tokyo Taro”, “Tokyo ...”, etc. The values are stored in association with each other (for example, the “name” of an individual whose “ID” is “1” is “Taro Tokyo” and “address” is “Tokyo ...”) Is associated).

  By the way, such a database is used in various usage forms. As one of the usage forms, for example, an individual specified by “data relating to an individual” stored in the database and a predetermined individual are the same person. There is a usage mode in which whether or not there is a possibility is collated and the result of collation is output. That is, since “data related to an individual” stored in the database is data corresponding to a specific individual, “data related to an individual” corresponding to a plurality of specific individuals is usually stored in the database. It will be. Therefore, by collating a plurality of “individual data” stored in the database with the data of a predetermined individual, the plurality of specific individuals stored in the database and the predetermined individual may be the same person It is possible to check whether or not there is a sex.

  For example, when there is a database that stores data corresponding to a plurality of customers before completion of repayment in a financial institution, for example, a predetermined customer who wishes to conclude a new contract before the completion of repayment In order for a financial institution to confirm whether or not the customer is a customer, it is possible to make a contract with a plurality of customers before completion of repayment stored in the database by comparing the data in the database with the data of this customer. It is verified whether or not this customer may be the same person.

  However, in reality, there are cases where the above-described usage pattern cannot be appropriately executed. In the example of the financial institution described above, this is a case where a predetermined customer who desires a contract impersonates another person using a forged driver's license. Normally, forged driver's licenses are completely prepared with fake "name" and fake "address", so it can be said that it is very difficult for financial institutions to detect that it is counterfeit information. . Even if you collate the data in the database with the completely forged "name" and "address" in this way, of course, with the same person as the customer who brought the forged driver's license and impersonated another person As a result, a collation result indicating that there is no possibility of data in the database is output.

  For this reason, for the purpose of combating such impersonation of another person, it is stored in a database using “personal biometric information” (for example, face image, fingerprint, vein, etc.) that is difficult to forge completely. Therefore, a “personal biometric information collating apparatus” that collates data related to an individual and “personal biometric information” related to a predetermined individual and outputs a result of the collation is required.

[Outline and Features of Personal Biometric Information Verification Device According to Embodiment 1]
Next, the outline and characteristics of the personal biometric information collating apparatus according to the first embodiment will be described with reference to FIG. FIG. 1 is a diagram for explaining the outline and features of the personal biometric information matching apparatus according to the first embodiment.

  As described above, the personal biometric information collation apparatus according to the first embodiment collates data related to individuals in a database that stores data relating to individuals classified by individual and personal biometric information related to a predetermined individual. The main feature is that the result of collation is output and the collation using personal biometric information is executed in a sophisticated and appropriate manner.

  This main feature will be briefly described. The personal biometric information collating apparatus according to the first embodiment holds a database that stores the feature amount of personal biometric information (a face image obtained by capturing a face) as personal data. (See (1) in (A) of FIG. 1). Note that the database according to the first embodiment includes a facial image feature amount calculated from a face image described in a publicly issued ID card, and a face image of an individual who is a subject. Are classified into personal feature amounts calculated from the captured face image and stored.

  Here, the feature amount of the face image stored in the database in the first embodiment is, for example, a collection of feature amounts such as pixel shading and change rate of the shading value in a local region around the sample point in the face image. Is a feature vector or the like expressed as a vector composed of real elements, and is a feature quantity that identifies a face after recognizing what kind of face the face image is.

  The personal biometric information collation apparatus configured as described above first calculates the feature amount of the personal biometric information (face image) from the personal biometric information (face image) related to a predetermined individual ((A) in FIG. 1). (See (2)). The personal biometric information collating apparatus according to the first embodiment further includes a scanner that captures an identification card issued by a public institution in which a face image is described, a camera that captures an individual's face image as a subject, and the like. A face image related to a predetermined individual is acquired by these scanners and cameras. In FIG. 1, for the sake of convenience, the face image is represented by an illustration, but it is actually image data captured by a scanner or camera and stored in a format such as JPEG (Joint Photographic Experts Group).

  Next, the personal biometric information matching device compares the feature quantity of the personal biometric information (face image) with the feature quantity of the personal biometric information (face image) stored in the database, A collation score indicating the degree of collation is calculated (see (3) in (B) of FIG. 1). As described above, the database according to the first embodiment classifies and stores the facial image feature quantities into identification feature quantities and personal feature quantities. When the feature value is calculated from the face image of the ID card imaged by a scanner, etc., the device compares this feature value with the ID card-based feature value of the database, calculates the matching score, and the camera When the feature amount is calculated from the face image of the individual captured by the above, a comparison score is calculated by comparing the feature amount with the personal feature amount of the database.

  And a personal biometric information collation apparatus outputs the result of collation determined by the calculated collation score to a predetermined output part (refer to (4) of (C) of Drawing 1). Note that the personal biometric information device according to the first embodiment corresponds to the calculated matching score (for example, 0.65, 0.41, etc.) and the feature amount compared when calculating the matching score as a result of matching. The face images to be associated with each other are output to the monitor in descending order of the matching score values. In addition, the personal biometric information collation apparatus according to the first embodiment outputs a collation result to a monitor in a form processed using a gauge that visually expresses a collation score value.

  In this way, the personal biometric information collation apparatus according to the first embodiment stores, for example, the feature amount calculated from the personal biometric information of a customer who wishes to conclude a contract (for example, the entire customer face image) and the database in advance. Using the feature quantity calculated from the stored customer's personal biometric information before completion of repayment, check whether the customer who wants to conclude a contract may be the same person as the customer before completion of repayment Therefore, compared with a method using part image data obtained by dividing face image data, collation using personal biometric information can be executed at a high level and appropriately.

  By the way, in addition to the main features described above, the personal biometric information collating apparatus according to the first embodiment has a feature amount of a face image calculated from a face image of an identification card imaged by a scanner and a face imaged by a camera. It is also characterized in that an identification confirmation collation score indicating the degree of collation between personal biometric information is calculated by comparing with the feature amount of the face image calculated from the image. In addition, the camera included in the personal biometric information collating apparatus according to the first embodiment continuously captures individual face images from the start of imaging until a predetermined time elapses, and the personal biometric information collating according to the first embodiment. The apparatus is also characterized in that an identity confirmation collation score is calculated only when a front face image is captured. In addition, in Example 1, although the case where an identification confirmation collation score is calculated only when a front face image is imaged is demonstrated, this invention is not limited to this, All the imaged faces are imaged. The present invention can be similarly applied to the case where an identification confirmation collation score is calculated from an image.

[Configuration of Personal Biometric Information Verification Device According to Embodiment 1]
Next, the configuration of the personal biometric information collating apparatus according to the first embodiment will be described with reference to FIGS. FIG. 2 is a block diagram illustrating the configuration of the personal biometric information collating apparatus according to the first embodiment, FIG. 3 is a diagram for explaining the output unit, and FIG. 4 illustrates the identification image capturing data holding unit. FIG. 5 is a diagram for explaining a personal imaging data holding unit, FIG. 6 is a diagram for explaining an identity confirmation collation score holding unit, and FIG. FIG. 8 is a diagram for explaining a confirmation score holding unit, FIG. 8 is a diagram for explaining a database holding unit, and FIG. 9 is a diagram for explaining a personal imaging data processing unit.

  As shown in FIG. 2, the personal biometric information matching device 10 according to the first embodiment includes an input unit 11, an identification image capturing unit 12, a personal image capturing unit 13, an output unit 14, and an input / output control IF unit 15. And a storage unit 20 and a control unit 30.

  The input unit 11 is means for inputting operation instructions for performing various processes by the control unit 30 by using a keyboard, a mouse, a communication unit, or the like. Specifically, the input unit 11 operates an output screen output by the output unit 14. For example, the input unit 11 displays an icon of an output screen output by the output unit 14 (for example, a “license scan” icon in FIG. 3) or a scroll bar (for example, a face image as a result of matching). , Etc.).

  The identity card imaging unit 12 is a means for imaging an identity card issued by a public institution in which personal biometric information is described. Specifically, the identification image capturing unit 12 captures an identification card in which personal biometric information is described, and the captured data is processed by an identification image capturing data processing unit 31a described later, and then described later. Is stored in the identification image pickup data holding unit 21a. For example, the identification image capturing unit 12 captures, with a scanner or the like, an identification document issued by a public institution in which face images are described, such as a driver's license, a passport, and a basic resident register card. The identification image capturing unit 12 corresponds to “identification image capturing means” described in the claims.

  The personal imaging unit 13 is a means for imaging personal biometric information of an individual who is a subject. Specifically, the personal imaging unit 13 images personal biometric information of an individual who is a subject, and the captured data is processed by a personal imaging data processing unit 31b described later, and then a personal imaging data holding unit 21b described later. Is remembered. For example, the personal imaging unit 13 captures a personal face image as a subject with a camera (digital camera) or the like. Here, the personal imaging unit 13 according to the first embodiment continuously captures personal face images until a predetermined time has elapsed since the start of imaging, and the front face image is captured by the personal imaging data processing unit 31b. When detected, the captured front face image is stored in the personal imaging data holding unit 21b. The personal imaging unit 13 corresponds to a “personal imaging unit” recited in the claims.

  The output unit 14 is a means for outputting the results of various processes by the control unit 30 and operation instructions for performing various processes to a monitor or a printer. Specifically, the output unit 14 outputs the collation result output by the later-described identity confirmation collation result output unit 33b, the collation result output by the duplication confirmation collation result output unit 34b described later, and the like. For example, the output unit 14 uses, as a result of the collation output by the identification confirmation collation result output unit 33b, a combination of an identification confirmation collation score and a face image calculated by an identification confirmation collation score calculation unit 33a described later on the monitor. Output. Further, for example, the output unit 14 compares the collation score calculated by the duplication confirmation collation score calculation unit 34a described later with the collation score calculated as the collation result output by the duplication confirmation collation result output unit 34b. The face images are associated with each other, and output to the monitor in descending order of the matching score values. For example, the output unit 14 outputs the value of the matching score to the monitor in a form processed using a visually expressing gauge.

  Here, an example of the output screen output by the output unit 14 will be described with reference to FIG. 3. The portion illustrated on the left side of the upper part of FIG. 3 is held by an identification image capturing data holding unit 21 a described later. A face image (upper) and a face image (lower) held by a personal imaging data holding unit 21b described later. When the icon output from the output unit 14 is operated by the input unit 11, the image capturing by the identification image capturing unit 12 is performed, or the image capturing by the personal image capturing unit 13 is started or stopped. In FIG. 3, a driver's license is illustrated as an identification card held by the identification card image data holding unit 21a. For example, a face image such as a passport or a basic resident register card is described. As long as it is issued by a public institution.

  Also, the part shown on the right side of the upper part of FIG. 3 is a collation result output by an identity confirmation collation result output unit 33b described later. In FIG. 3, “0.81” is used as the identification confirmation collation score calculated by the later-described identification confirmation collation score calculation unit 33 a, and the face image (in the personal imaging data holding unit 21 b) compared when calculating the identification confirmation collation score. In addition to the combination with (held face image), the face image held by the identification image pickup data holding unit 21a is also illustrated. The status bar of “elapsed time” indicates the elapsed time from the start of imaging until a predetermined time elapses when an individual face image is continuously captured by the individual imaging unit 13. Is displayed.

  Further, the part shown in the lower part of FIG. 3 is a collation result output by the duplication confirmation collation result output unit 34b described later. In FIG. 3, the matching score calculated by the duplication check matching score calculation unit 34a is associated with the face image (face image held by the database holding unit 25) compared at the time of calculating the matching score, and the matching score is calculated. The values are shown in descending order.

  Here, in FIG. 3, the face image captured by the ID image capturing unit 12 (scanner) and the face image of the ID card system held by the database holding unit 25 (the face of the ID card imaged by the scanner or the like). The result of the image matching is shown in the upper row, and the face image captured by the personal imaging unit 13 (camera) and the personal face image held by the database holding unit 25 (subject captured by the camera or the like) The result of the comparison with the individual face image is shown in the lower row. Further, FIG. 3 illustrates a gauge that visually represents the value of the matching score between the matching score and the face image.

  The input / output control IF unit 15 is means for controlling data transfer among the input unit 11, the identification image capturing unit 12, the personal image capturing unit 13, and the output unit 14, and the storage unit 20 and the control unit 30.

  The storage unit 20 is a storage unit that stores data used for various types of processing by the control unit 30. In particular, as closely related to the present invention, as shown in FIG. The holding part 22, the identification confirmation collation score holding part 23, the duplication confirmation collation score holding part 24, and the database holding part 25 are provided. The database holding unit 25 corresponds to “database holding means” described in the claims.

  In the storage unit 20, the imaging data holding unit 21 is a unit that holds personal biometric information related to a predetermined individual that has been imaged, and includes an identification card imaging data holding unit 21 a, a personal imaging data holding unit 21 b, and the like. Is provided.

  The identification image capturing data holding unit 21a is a means for holding the personal biometric information described in the imaged ID issued by a public institution. Specifically, the identification image capturing data holding unit 21 a holds personal biometric information described in a public institution-issued identification certificate imaged by the identification certificate imaging unit 12. For example, as shown in FIG. 4, the ID card image data holding unit 21a holds a face image described in a driver's license imaged by a scanner or the like. Further, the data held by the identification image capturing data holding unit 21a includes a feature amount calculation process by a feature amount calculation unit 32, which will be described later, an identification confirmation collation result output process, which is described later, and a duplication, which will be described later. This is used for the duplication confirmation collation result output processing by the certain collation result output unit 34b.

  The personal imaging data holding unit 21b is means for holding personal biometric information of an individual who is an imaged subject. Specifically, the personal imaging data holding unit 21 b holds personal biometric information of an individual who is a subject imaged by the personal imaging unit 13. For example, as shown in FIG. 5, the personal imaging data holding unit 21b holds a personal face image captured by a camera. Here, the personal image data holding unit 21b according to the first embodiment detects a front face image detected by a personal image data processing unit 31b described later from among a plurality of face images continuously captured by the personal image capturing unit 13. Only the face image of the case is retained. The data held by the personal imaging data holding unit 21b is used for a feature amount calculation process by a feature amount calculation unit 32, which will be described later, a duplication confirmation collation result output process by a duplication confirmation result output unit 34b, which will be described later.

  The feature amount holding unit 22 is means for holding the calculated feature amount of the personal biometric information, and includes an identification card feature amount holding unit 22a and a personal feature amount holding unit 22b.

  The identification card feature value holding unit 22a is a unit that holds the feature value of the personal biometric information calculated by the feature value calculation unit 32 (to be described later) from the personal biometric information held by the identification image pickup data holding unit 21a. . Further, the feature amount held by the identification certificate feature amount holding unit 22a is used for an identification confirmation collation score calculation process by an identification confirmation collation score calculation unit 33a described later and a duplication confirmation collation score calculation process by a duplication confirmation collation score calculation unit 34a. Is done.

  The personal feature amount holding unit 22b is means for holding the feature amount of the personal biometric information calculated by the feature amount calculating unit 32 described later from the personal biometric information held by the personal imaging data holding unit 21b. Here, the personal feature amount holding unit 22b according to the first embodiment detects a front face image by a personal imaging data processing unit 31b described later from among a plurality of face images continuously captured by the personal imaging unit 13. Each time, the feature amount of the face image calculated by the feature amount calculation unit 32 is held. Further, the feature amount held by the individual feature amount holding unit 22b is used for an identification confirmation collation score calculation process by an identification confirmation collation score calculation unit 33a described later and a duplication confirmation collation score calculation process by the duplication confirmation collation score calculation unit 34a. .

  The identification confirmation collation score holding unit 23 is means for holding the calculated identification confirmation collation score. Specifically, the identification confirmation collation score holding unit 23 includes the feature quantity of the personal biometric information held by the identification certificate feature quantity holding unit 22a and the feature quantity of the personal biometric information held by the personal feature quantity holding unit 22b. The identification confirmation collation score calculated by the identification confirmation collation score calculation unit 33a is held. Further, the identification confirmation collation score held by the identification confirmation collation score holding unit 23 is used for an identification confirmation collation result output process by an identification confirmation collation result output unit 33b described later.

  Here, an example of the form in which the identification confirmation collation score holding unit 23 holds the identification confirmation collation score will be described with reference to FIG. As described above, the personal feature amount holding unit 22b according to the first embodiment detects a front face image by the personal imaging data processing unit 31b described later from among a plurality of face images continuously captured by the personal imaging unit 13. Each time, the feature amount of the face image calculated by the feature amount calculation unit 32 is held. Therefore, the identity confirmation collation score calculation unit 33a in the first embodiment described later calculates a plurality of identity confirmation collation scores according to the plurality of front face images. For this reason, as shown in FIG. 6, the identification confirmation collation score holding unit 23 according to the first embodiment retains a plurality of identification confirmation collation scores, and also compares the front faces of individuals compared in calculating the identification confirmation collation score. Keep a combination of images. Furthermore, the identification confirmation collation score holding unit 23 according to the first embodiment includes an identification confirmation collation score having the highest value among the plurality of identification confirmation collation scores (best), an identification confirmation collation score having the lowest value (worst), The average value calculated by the identification confirmation collation score calculation unit 33a is held.

  The duplication confirmation collation score holding unit 24 is a means for holding the calculated collation score. Specifically, as shown in FIG. 7, the duplication confirmation collation score holding unit 24 includes the feature quantity of the personal biometric information held by the identification card feature quantity holding unit 22 a and the identification card held by the database holding unit 25. A collation score calculated by a duplication confirmation collation score calculation unit 34a, which will be described later, is stored from the typeface feature amount, and is stored by the database storage unit 25 and the feature amount of the personal biometric information held by the personal feature amount holding unit 22b. The collation score calculated by the duplication confirmation collation score calculation unit 34a described later is stored from the individual feature amount. The collation score held by the duplication confirmation collation score holding unit 24 is used for duplication confirmation collation result processing by the duplication confirmation collation result output unit 34b described later.

  The database holding unit 25 is a database that stores data related to individuals by classifying the data for each individual, and is a means for holding a database that stores the feature amount of the personal biometric information of the individual as the data related to the individual. Specifically, the database holding unit 25 holds a database that stores the feature amount of the personal face image as data relating to the individual. Here, the database held by the database holding unit 25 according to the first embodiment includes the feature amount of the face image, the identification-based feature amount calculated from the face image described in the captured identification card, and the subject. Are classified into personal feature amounts calculated from the face images obtained by capturing the individual face images.

  Here, an example of a form in which the database held by the database holding unit 25 stores personal data will be described with reference to FIG. 8. The database in the first embodiment includes attributes “ID”, “ Specific values such as “name”, “address”, “identification face image”, “identification feature value”, “personal face image”, and “personal feature value” for each individual Separate and memorize. Note that the “identification feature value” and the “personal feature value” are feature values calculated by the same technique as that used by the feature value calculation unit 32 described later.

  The control unit 30 is means for executing the various processes by controlling the personal biometric information collating apparatus 10, and particularly as closely related to the present invention, as shown in FIG. A feature amount calculation unit 32, an identity confirmation unit 33, and an overlap confirmation unit 34 are provided.

  In the control unit 30, the imaging data processing unit 31 is a means for processing captured data, and includes an identification card imaging data processing unit 31a and a personal imaging data processing unit 31b.

  The identification card image data processing unit 31a is a unit that processes the imaging data of the identification card imaged by the identification card imaging unit 12. Specifically, the identification card imaging data processing unit 31a performs skew correction of the imaging data of the identification card, specifies the position of the face image of the identification card imaging data subjected to the skew correction, and detects the facial image. The face image is extracted from the ID card image data specifying the position of the ID card, and the extracted face image is stored in the ID card image data storage unit 21a.

  The personal imaging data processing unit 31b is a means for processing the imaging data of the facial image captured by the personal imaging unit 13, and when the personal facial image that is the subject is continuously captured, the personal front face Means for detecting an image. Specifically, the personal imaging data processing unit 31b detects the personal front image when the personal imaging unit 13 continuously captures individual face images until a predetermined time elapses after the imaging starts. The face image is detected, and the detected front face image is stored in the personal imaging data holding unit 21b.

  For example, as shown in FIG. 9, the individual imaging data processing unit 31 b moves the individual who is the subject while the imaging is continuously performed until a predetermined time elapses after the imaging is started by the individual imaging unit 13. As a result, face images such as “face facing upward”, “face facing left”, “face facing front”, “face facing right”, and “face facing downward” are captured. If it has been done, only the face image facing the front is detected and stored in the personal imaging data holding unit 21b.

  The feature amount calculation unit 32 is a means for calculating the feature amount of the personal biometric information from the personal biometric information related to a predetermined individual. Specifically, the feature amount calculation unit 32 calculates the feature amount of the personal biometric information from the personal biometric information held by the identification image pickup data holding unit 21a and stores it in the identification card feature amount holding unit 22a. . In addition, the feature amount calculation unit 32 calculates the feature amount of the personal biometric information from the personal biometric information held by the personal imaging data holding unit 21b and stores it in the personal imaging data holding unit 22b. The feature quantity calculation unit 32 corresponds to “feature quantity calculation means” described in the claims.

  Here, the feature amount calculation unit 32 in the first embodiment uses a “feature vector” as the feature amount of the face image calculated from the face image. A “feature vector” is a collection of feature quantities such as pixel shading in the local area around the sample points in the face area (for example, 100 arbitrary points in the face area) and the change rate of the shading value. Is expressed as a vector composed of real elements.

  Further, the concept of “subspace” used in the feature quantity calculation unit 32 will be described. First, in the feature vector space, there are a “substantial variation subspace” and a “others variation subspace”. The "substantial variation subspace" is a subspace composed of differences in feature vectors in the image data of different individuals depending on differences in facial expressions, shooting angles, etc. This is a partial space composed of the difference of feature vectors in the image data pair of the person and the other person. A space where the complementary space of the “person's fluctuation subspace” and the “other person's fluctuation subspace” intersect is called a “recognition subspace”.

  The feature amount calculation unit 32 according to the first embodiment calculates the distance to the “recognition subspace” of the feature vector in the face area, and detects sample points on the face area such that the distance becomes small. When the sample point is detected, the feature amount calculation unit 32 creates a new recognition subspace based on the feature vector of the new sample point in the face area. The feature amount calculation unit 32 detects the sample point based on the distance of the feature vector to the recognition subspace until the created recognition subspace converges, and the recognition subspace based on the feature vector at the new sample point. Repeat the creation process. In this way, the feature amount calculation unit 32 calculates the feature vector with high accuracy. In the first embodiment, the method in which the feature amount calculation unit 32 calculates the feature amount by the above-described method has been described. However, the present invention is not limited to the above-described method, and the feature amount of the face image is calculated. Any method may be used as long as it is a calculation method.

  The identity confirmation unit 33 is a means for collating the personal biometric information captured by the identification image capturing unit 12 with the personal biometric information captured by the personal image capture unit 13, and includes an identity confirmation verification score calculation unit 33a and an identity confirmation. And a collation result output unit 33b. The identification confirmation collation score calculation unit 33a corresponds to the “identification confirmation collation score calculation unit” recited in the claims, and the identification confirmation collation result output unit 33b includes the “output unit” recited in the claims. Corresponding to

  The identification confirmation collation score calculation unit 33a is a feature amount of the personal biometric information calculated from the personal biometric information captured by the identification image capturing unit 12 and an individual calculated from the personal biometric information captured by the personal imaging unit 13. This is a means for comparing the feature quantity of the biometric information and calculating an identification confirmation collation score indicating the degree of collation between the individual biometric information. Specifically, the identification confirmation collation score calculation unit 33a includes the feature quantity of the personal biometric information held by the identification certificate feature quantity holding unit 22a and the feature quantity of the personal biometric information held by the personal feature quantity holding unit 22b. Are compared with each other to calculate an identification confirmation collation score indicating the degree of collation between the personal biometric information, and store it in the identification confirmation collation score holding unit 23. In addition, when the personal imaging unit 13 captures a plurality of individual face images, the identification confirmation collation score calculation unit 33a according to the first embodiment calculates a plurality of identification confirmation collation scores according to the plurality of face images. An average value of the plurality of identification confirmation collation scores is calculated.

  Here, as described above, the personal feature amount holding unit 22b according to the first exemplary embodiment has the front face image captured by the personal imaging data processing unit 31b among the plurality of face images continuously captured by the personal imaging unit 13. Each time it is detected, the feature amount of the face image calculated by the feature amount calculation unit 32 is held. Accordingly, the identification confirmation collation score calculation unit 33a calculates a plurality of identification confirmation collation scores according to the plurality of front face images, and the identification confirmation collation score holding unit 23, as shown in FIG. The collation score is stored, and the face images of the front faces of the individuals compared in calculating the identification confirmation collation score are stored in combination. Furthermore, the identification confirmation collation score calculation unit 33a in the first embodiment causes the identification confirmation collation score holding unit 23 to have the highest value of the plurality of identification confirmation collation scores (best) and the lowest value. The identification confirmation collation score (worst) and the average value calculated by the identification confirmation collation score calculation unit 33a are stored.

  The identification confirmation collation result output unit 33b is a unit that outputs the identification confirmation collation score calculated by the identification confirmation collation score calculation unit 33a to the output unit 14. Specifically, the identification confirmation collation result output unit 33b combines the identification confirmation collation score held by the identification confirmation collation score holding unit 23 with the face image of the front of the individual compared when calculating the identification confirmation collation score. Output to a predetermined output unit. For example, the identification confirmation collation result output unit 33b outputs the collation result to a monitor or the like as shown in the right part of FIG. In Example 1, the identification confirmation collation result output unit 33b was compared in calculating the highest value identification confirmation collation score and the highest value identification confirmation collation score held by the identification confirmation collation score holding unit 23. A combination of face images of the front of an individual is output to a monitor or the like, but the present invention is not limited to this, and the comparison was made when calculating the lowest value identification verification score and the lowest value identification verification score. Any one of the combinations of face images of the front of the individual, the average value of the identification confirmation collation score, and a predetermined display indicating that the identification confirmation collation score has exceeded the predetermined threshold when the identification confirmation collation score exceeds the predetermined threshold Alternatively, a plurality may be output.

  The duplication confirmation unit 34 is a means for collating the personal biometric information stored in the database held by the database holding unit 25 with the personal biometric information related to a predetermined individual. The duplication confirmation collation score calculation unit 34a and the duplication confirmation And a collation result output unit 34b. The duplication confirmation collation score calculation unit 34a corresponds to the “collation score calculation unit” described in the claims, and the duplication confirmation collation result output unit 34b corresponds to the “output unit” described in the claims. To do.

  The duplication confirmation collation score calculation unit 34a compares the feature quantity of the personal biometric information calculated from the personal biometric information related to the predetermined individual with the feature quantity of the personal biometric information stored in the database, and It is a means for calculating a matching score indicating the degree of matching. Specifically, the duplication confirmation collation score calculation unit 34a calculates the feature amount of the personal biometric information held by the identification certificate feature amount holding unit 22a and the identification type feature amount held by the database holding unit 25. The comparison score is calculated by comparison, and the comparison score is calculated by comparing the feature quantity of the personal biometric information held by the personal feature quantity holding unit 22b with the personal feature quantity held by the database holding unit 25. And stored in the duplication confirmation collation score holding unit 24.

  The duplication confirmation collation result output unit 34 b is a unit that outputs the collation result determined by the collation score to the output unit 14. Specifically, the duplication confirmation collation result output unit 34b is held by the database holding unit 25 which is compared with the collation score calculated by the duplication confirmation collation score calculation unit 34a and the collation score calculation as a collation result. The corresponding face image is output to the monitor. For example, as shown in the lower part of FIG. 3, the duplication confirmation collation result output unit 34b compares the collation score calculated by the duplication confirmation collation score calculation unit 34a with the face image (database holding) compared with the collation score calculation. The face images held by the unit 25 are associated with each other and output to the monitor in descending order of the matching score values.

  Further, as shown in the lower part of FIG. 3, the duplication confirmation collation result output unit 34 b includes the face image captured by the identification image capturing unit 12 (scanner) and the identification system stored by the database storage unit 25. The face image (the face image of the ID card imaged by a scanner or the like) is collated with the face image captured by the personal imaging unit 13 (camera) and held by the database holding unit 25. The result of collation with the personal face image (personal face image as a subject imaged by a camera or the like) is output to the lower row. Further, the duplication confirmation collation result output unit 34b outputs a gauge that visually represents the collation score value between the collation score and the face image.

[Procedure for Processing by Personal Biometric Information Collating Device According to Embodiment 1]
Next, a processing procedure by the personal biometric information collating apparatus according to the first embodiment will be described with reference to FIGS. FIG. 10 is a flowchart illustrating a procedure of processing performed by the personal biometric information collating apparatus according to the first embodiment. FIG. 11 is a flowchart illustrating a procedure of identification card imaging data processing. FIG. FIG. 13 is a flowchart illustrating the procedure of the calculation process, and FIG. 13 is a flowchart illustrating the procedure of the calculation process of the overlap confirmation matching score.

  Here, the personal biometric information collation apparatus 10 according to the first embodiment describes the feature amount of the personal biometric information (a face image obtained by capturing a face) of the individual as data related to the individual in an identification document issued by a public institution. A database is stored that categorizes and stores personal identification features calculated from facial images calculated from facial images captured from the facial images of individuals and personal facial features calculated from the captured facial images of individuals. Part 25 is held. In addition, the personal biometric information matching device 10 according to the first embodiment checks the database holding unit 25 to check whether there is any duplication of face images (matrix matching for cross-matching all face images in the database). ) Are detected in advance and a duplicate face image is detected in advance and an alarm or flag is set. In the following, first, the procedure of the entire personal biometric information matching process by the personal biometric information matching apparatus 10 holding such a database will be described, and then the procedure of the identification certificate imaging data process, the identification confirmation matching score calculation process will be described. The procedure and the procedure of the duplication confirmation collation score calculation process will be described in detail.

[Personal biometric information verification overall procedure]
First, the personal biometric information matching device 10 according to the first embodiment determines whether or not personal biometric information has been imaged by the identification image capturing unit 12 (step S1001). Specifically, the personal biometric information collation apparatus 10 according to the first embodiment uses an identification document issued by a public institution (for example, a face image is described) by an identification image capturing unit 12 (for example, a scanner). It is determined whether a face image of a driver's license or the like has been captured.

  When the face image of the identification card is not captured (No at Step S1001), the personal biometric information matching device 10 performs a process of determining whether the personal biometric information is captured by the identification card imaging unit 12. Return (step S1001). On the other hand, when the face image of the identification card is captured (Yes at step S1001), the personal biometric information matching device 10 next performs identification card imaging data processing (step S1002). Specifically, the personal biometric information collation apparatus 10 processes the data imaged by the identification image capturing unit 12, holds the face image in the identification image captured data holding unit 21a, or stores the face image of the held face image. The feature amount is calculated and the feature amount of the face image is held in the identification feature amount holding unit 22a.

  Subsequently, the personal biometric information collation apparatus 10 determines whether or not the personal imaging unit 13 has started imaging of personal biometric information (step S1003). Specifically, the personal imaging unit 13 (for example, a camera or the like) in the first embodiment continuously captures personal face images from the start of imaging until a predetermined time elapses. The collation apparatus 10 determines whether or not the personal imaging unit 13 has started imaging of a personal face image.

  When imaging of a personal face image has not been started (No at Step S1003), the personal biometric information collation apparatus 10 performs processing for determining whether imaging of personal biometric information has been started by the personal imaging unit 13. Return (step S1003). On the other hand, when imaging of a personal face image is started (Yes in step S1003), the personal biometric information collating apparatus 10 next determines whether or not a front face image is detected (step S1004). Specifically, in the personal biometric information collating apparatus 10 according to the first embodiment, when a personal front face image is captured by the personal imaging unit 13, a personal front face image is stored in the personal imaging data holding unit 21b. Therefore, the personal image data processing unit 31b determines whether or not a front face image is detected.

  When the front face image is not detected (No at Step S1004), the personal biometric information collation apparatus 10 returns to the process of determining whether or not the front face image is detected (Step S1004). On the other hand, if a front face image is detected (Yes at step S1004), the personal biometric information collation apparatus 10 next performs an identification confirmation collation score calculation process (step S1005). Specifically, the personal biometric information matching device 10 holds the front face image detected by the personal image data processing unit 31b in the personal image data holding unit 21b, or calculates the feature amount of the held face image. The feature amount of the face image is held in the individual feature amount holding unit 22b, or the identification confirmation collation score is calculated using the held feature amount.

  Then, the personal biometric information collation apparatus 10 determines whether or not a predetermined time has elapsed after the personal imaging unit 13 (for example, a camera or the like) has started capturing an individual's face image (step S1006). That is, as described above, the personal imaging unit 13 according to the first embodiment continuously captures individual face images from the start of imaging until a predetermined time elapses. If the predetermined time has not elapsed (No at Step S1006), the information collation apparatus 10 repeats the process of determining the detection of the front face image (Step S1004) and the identification confirmation collation score calculation process (Step S1005). .

  On the other hand, if the predetermined time has elapsed (Yes at step S1006), the personal biometric information collation apparatus 10 next performs an identification confirmation collation score calculation process (average value calculation) (step S1007). Specifically, as described above, the personal biometric information matching apparatus 10 according to the first embodiment performs a plurality of steps in step S1005 according to a plurality of face images captured by the personal imaging unit 13 (for example, a camera). Since the identification confirmation collation score is calculated, an average value of a plurality of identification confirmation collation scores is calculated in step S1007.

  And the personal biometric information collation apparatus 10 performs an identification confirmation collation result output process (step S1008). Specifically, the personal biometric information collation apparatus 10 outputs the identification confirmation collation score to the output unit 14 (for example, a monitor) in the identification confirmation collation result output unit 33b. In addition, the personal biometric information collation apparatus 10 according to the first embodiment combines a face image of the front of the individual compared when calculating the highest value identification check score and the highest value confirmation check score, and the like. Are output to the combination of the face image of the front of the individual compared when calculating the lowest value identification verification score and the lowest value identification verification score, or a plurality of identification verification scores calculated in step S1007 Since the average value is also held, these may be selected and output.

  Subsequently, the personal biometric information collation apparatus 10 determines whether or not a duplication confirmation instruction has been received (step S1009). That is, the personal biometric information collation apparatus 10 according to the first embodiment outputs the identification confirmation collation result to a monitor or the like in step S1008. The output identification confirmation collation result is obtained by operating the personal biometric information collation apparatus 10 here. It will be confirmed by the operator who performs. The personal biometric information collation apparatus 10 determines that the duplication confirmation is to be continued by an operator or the like, and starts the process related to duplication confirmation when the duplication confirmation instruction is input by the operator or the like. In step S1009, It is determined whether a duplicate confirmation instruction has been accepted.

  When the duplication confirmation instruction has not been accepted (No at Step S1009), the personal biometric information collation apparatus 10 returns to the process of determining whether or not the duplication confirmation instruction has been accepted (Step S1009). On the other hand, if a duplicate confirmation instruction is accepted (Yes at step S1009), the personal biometric information collation apparatus 10 next performs a duplicate confirmation collation score calculation process (step S1010). Specifically, the personal biometric information matching device 10 compares the facial image feature quantity held in the identification certificate feature quantity holding unit 22a with the identification card type feature quantity held in the database holding unit 25. Thus, the collation score is calculated, or the collation score is calculated by comparing the feature quantity held in the personal feature quantity holding unit 22b with the personal feature quantity held in the database holding unit 25, respectively.

  And personal biometric information collation apparatus 10 performs duplication check collation result output processing (Step S1011). Specifically, the personal biometric information collation apparatus 10 uses the duplication confirmation collation result output unit 34b to obtain the collation score and the face image held by the database holding unit 25 compared when calculating the collation score as a collation result. In association with each other, the data is output to the output unit 14 (for example, a monitor). The personal biometric information matching device 10 according to the first embodiment associates the matching score with the face images compared when calculating the matching score, and outputs the matching score to the monitor in descending order of the matching score value. A gauge that visually expresses the value of the matching score is output between the face image and the face image.

  In this way, the personal information collation apparatus 10 according to the first embodiment stores, for example, the feature amount calculated from the personal biometric information (for example, the entire customer face image) of the customer who wishes to conclude a contract in advance in the database Using the feature quantity calculated from the stored customer's personal biometric information before completion of repayment, check whether the customer who wants to conclude a contract may be the same person as the customer before completion of repayment Therefore, compared with a method using part image data obtained by dividing face image data, collation using personal biometric information can be executed at a high level and appropriately.

  In the first embodiment, the personal information collation apparatus 10 first performs the process related to identity confirmation, and then the process for performing the process related to duplication confirmation. However, the present invention is limited to this. It is not a thing, but a process related to duplication confirmation, such as a procedure for performing only duplication confirmation processing without performing identification confirmation processing (for example, a process for calculating a collation score, or a determination by a calculated collation score) As long as it is a procedure for performing a process for outputting the result of collation to a predetermined output unit), the presence / absence or order of the process for the identification confirmation may be any.

  In the first embodiment, the personal information collation apparatus 10 performs the process related to identity confirmation and then performs the process related to the duplicate confirmation through the confirmation by the operator or the like. This is not a limitation, and after the process related to identity confirmation, without confirmation by the operator, etc., the process procedure for performing the process related to duplicate confirmation, and the confirmation by the operator etc. after the duplicate confirmation collation result output process The presence / absence and timing of confirmation by the operator in the personal information matching process by the personal information matching device 10 may be any, such as the procedure of the process passing through.

[Procedure for processing ID data]
Next, the procedure of the ID card image data processing in step S1002 will be described in detail.

  First, the personal biometric information collation apparatus 10 takes in the identification image capturing data captured by the identification image capturing unit 12 in the identification image capturing data processing unit 31a (step S1101).

  Next, the personal biometric information collation apparatus 10 performs skew correction of the captured identification image imaging data in the identification certificate imaging data processing unit 31a (step S1102).

  Then, the personal biometric information collating apparatus 10 specifies the position of the face image of the identification image pickup data subjected to the skew correction in the identification certificate image pickup data processing unit 31a (step S1103).

  Subsequently, the personal biometric information collation apparatus 10 extracts the face image from the identification image capturing data specifying the position of the face image in the identification image capturing data processing unit 31a (step S1104).

  The personal biometric information collation apparatus 10 causes the identification image capturing data processing unit 31a to hold the extracted face image in the identification certificate imaging data holding unit 21a (step S1105).

  Next, the personal biometric information collation apparatus 10 calculates the feature amount of the face image held in the identification image pickup data holding unit 21a in the feature amount calculation unit 32 (step S1106). Specifically, the personal biometric information matching device 10 calculates the feature amount of the face image held in the identification image capturing data holding unit 21a in the feature amount calculation unit 32, and calculates the feature amount of the calculated face image. It is held in the identification card feature amount holding unit 22a.

  In the first embodiment, the case has been described in which the identification image capturing data processing unit 31a corrects the skew of the identification image capturing data captured by the identification image capturing unit 12, but the present invention is not limited thereto. The present invention is not limited, and any method may be used as long as the face image of the ID card is held in the ID card image data holding unit 21a, such as when the position of the face image is specified without performing skew correction. . Further, in the first embodiment, the case where the personal biometric information matching device 10 calculates the feature amount of the face image immediately after holding the face image of the identification card in the identification card imaging data holding unit 21a has been described. However, the present invention is not limited to this, and the timing of calculating the feature amount of the face image is, for example, when calculating the feature amount of the face image of the identification card in the identification confirmation collation score calculation processing in step S1005. Any of them may be used.

[Identification verification verification score calculation procedure]
Next, the procedure of the identification confirmation collation score calculation process in step S1005 described above will be described in detail.

  First, the personal biometric information collating apparatus 10 holds the front face image detected by the personal imaging data processing unit 31b in the personal imaging data holding unit 21b (step S1201).

  Then, the personal biometric information collation apparatus 10 calculates the feature amount of the face image held in the personal imaging data holding unit 21b in the feature amount calculation unit 32 (step S1202). Specifically, in the personal biometric information collating apparatus 10, the feature amount calculation unit 32 calculates the feature amount of the face image held in the personal imaging data holding unit 21b, and the calculated feature amount of the face image is used as the personal feature. The amount is held in the amount holding unit 22b.

  Subsequently, the personal biometric information collation apparatus 10 compares the feature amount held in the identification certificate feature amount holding unit 22a and the feature amount held in the personal feature amount holding unit 22b in the identification collation score calculation unit 33a. Then, an identification check collation score indicating the degree of collation between the personal biometric information is calculated (step S1203).

  The personal biometric information collation apparatus 10 holds the combination of the calculated identification confirmation collation score and the front face image compared in the calculation in the identification confirmation collation score holding unit 23 in the identification confirmation collation score calculation unit 33a. (Step S1204).

[Duplication confirmation collation score calculation processing procedure]
Next, the procedure of the duplication confirmation collation score calculation process in step S1010 described above will be described in detail.

  First, the personal biometric information collation apparatus 10 uses the duplication confirmation collation score calculation unit 34a to store the feature amount held in the identification certificate feature amount holding unit 22a and the identification type feature amount held in the database holding unit 25. Are compared to calculate a matching score (step S1301). Specifically, the personal biometric information matching device 10 compares and compares the feature quantity held in the identification feature quantity holding unit 22a with the identification type feature quantity in the duplication confirmation matching score calculation unit 34a. The score is calculated, and the calculated matching score is held in the duplication check matching score holding unit 24.

  Then, the personal biometric information collation apparatus 10 compares the feature amount held in the personal feature amount holding unit 22b with the personal feature amount held in the database holding unit 25 in the duplication confirmation collation score calculation unit 34a. The collation score is calculated (step S1302). Specifically, the personal biometric information collation apparatus 10 calculates a collation score by comparing the feature amount held in the personal feature amount holding unit 22b with the individual feature amount in the duplication confirmation collation score calculation unit 34a. The calculated matching score is held in the duplication check matching score holding unit 24.

  In the first embodiment, the personal information collation apparatus 10 calculates a collation score by comparing the identification system feature amount and the like in the duplication confirmation collation score calculation unit 34a, and then compares the personal feature amount and the like. In the duplication confirmation collation result output unit 34b, the procedure for outputting the collation result has been described. However, the present invention is not limited to this, and the duplication confirmation collation score calculation unit In 34a, a collation score that compares identification feature quantities and the like is calculated, and after the collation confirmation collation result output section 34b outputs the collation result, the duplication confirmation collation score calculation section 34a performs personal system characteristic quantities and the like. That outputs a collation result, such as a processing procedure for outputting a collation result in the duplication confirmation collation result output unit 34b. And the like order, may be either.

[Effect of Example 1]
As described above, according to the invention of the first embodiment, data related to individuals in a database that stores data related to individuals classified and stored with personal biometric information related to a predetermined individual is verified. The personal biometric information collating device that outputs the result of the above, which holds a database that stores the feature quantity of the personal biometric information of the individual as the data related to the individual, and the feature quantity of the personal biometric information from the personal biometric information of the predetermined individual Is calculated, and the feature quantity of the personal biometric information calculated is compared with the feature quantity of the personal biometric information stored in the stored database, and a collation score indicating the degree of collation between the personal biometric information is calculated and calculated. Since the result of the collation determined by the collation score is output to a predetermined output unit, for example, from the personal biometric information (for example, the entire customer face image, etc.) The customer who wants to conclude a contract is the same person as the customer before the completion of the repayment using the feature amount to be issued and the feature amount calculated from the personal biometric information of the customer before the completion of the repayment stored in the database in advance. Compared to the method using part image data obtained by dividing face image data, it is possible to execute verification using personal biometric information in a sophisticated and appropriate manner because it is a method for checking whether or not there is a possibility. It becomes possible.

  In addition, according to the invention of the first embodiment, a database that stores the feature amount of a face image obtained by capturing a face is stored as the feature amount of the personal biometric information, and the feature amount of the face image is used as the feature amount of the personal biometric information. Since the comparison score is calculated by comparing the calculated feature value of the face image with the feature value of the face image stored in the database, the method of matching based on the feature value of the face image obtained by capturing the face is: Compared to the method of collating based on the feature quantity of other personal biometric information such as fingerprint feature quantity and vein feature quantity, it is easier to collect the predetermined data used for calculating the feature quantity. Collation using information can be executed in an advanced, appropriate and simple manner.

  Further, according to the invention of the first embodiment, since an identity certificate issued by a public institution in which personal biometric information is described is imaged, and the feature quantity of the personal biometric information is calculated from the captured personal biometric information, for example, Since the face image used for calculating the feature amount of the face image is obtained by capturing an ID issued by a public institution (eg, driver's license, passport) with a scanner, etc. Collation using information can be executed in an advanced, appropriate and simple manner.

  Further, according to the invention of the first embodiment, the personal biometric information of the individual who is the subject is imaged, and the feature amount of the personal biometric information is calculated from the captured personal biometric information. Since the face image used in the above is acquired by capturing an individual's face as a subject with a camera or the like, collation using personal biometric information can be executed in an advanced, appropriate and simple manner.

  Further, according to the invention of the first embodiment, the stored database includes the feature amount of the personal biometric information and the identification-based feature amount calculated from the personal biometric information described in the captured identification card. The personal biometric information of the individual who is the subject is classified and stored in the personal system feature amount calculated from the captured personal biometric information, and from the personal biometric information described in the captured identification card When the feature amount of the personal biometric information is calculated, the feature score is compared with the feature value of the identification card system to calculate the matching score, and the feature amount of the personal biometric information is calculated from the captured personal biometric information of the individual In this case, since the comparison score is calculated by comparing the feature quantity with the personal feature quantity, for example, the feature quantities calculated from the face photograph of the driver's license taken with a scanner or the like, or the camera, etc. Individuals imaged at The method of calculating the matching score by comparing the feature quantities calculated from the facial photos reduces the difference in the environment such as the imaging conditions, for example, calculated from the facial photo of the driver's license taken with a scanner etc. Compared with the technique of calculating the matching score by comparing the feature quantity with the feature quantity calculated from the personal face photograph taken with a camera etc., matching using personal biometric information is advanced, appropriate and accurate It becomes possible to execute.

  In addition, according to the invention of the first embodiment, the feature amount of the personal biometric information calculated from the personal biometric information described in the captured identity card and the individual calculated from the captured personal biometric information of the individual Before comparing the data related to individuals in the database with the personal biometric information related to a predetermined individual, since the identification check score indicating the degree of matching between the personal biometric information is calculated by comparing with the feature amount of the biometric information. As a stage, for example, a face photo of a driver's license taken with a scanner or the like and a personal face photo taken with a camera or the like are collated to confirm the identity (identity confirmation), and a predetermined individual to be collated Since it is confirmed that the personal biometric information related to the personal data is the person's data, it is possible to execute collation using the personal biometric information at a high level and more appropriately.

  Further, according to the invention of the first embodiment, as a result of the collation, the calculated collation score and the data relating to the individual compared in the calculation of the collation score are output in association with each other. A collation score and a face image are output in correspondence with a predetermined output unit (monitor), and the result of the collation is presented to an operator or the like, so that collation using personal biometric information is advanced, appropriate and suitable for operation. It becomes possible to execute with.

  Further, according to the invention of the first embodiment, the collation results are arranged and output to a predetermined output unit in descending order of the calculated collation score value. For example, the collation score is output to a predetermined output unit (monitor). Face images are arranged and output in descending order, and verification is performed from the highest collation score value by an operator or the like, so that collation using personal biometric information can be executed at an advanced, appropriate and efficient level. Become.

  Further, according to the invention of Example 1, since the result of matching is output to a predetermined output unit in a form processed using a gauge that visually expresses the value of the calculated matching score, the value of the matching score Therefore, it is possible to perform collation using personal biometric information in a sophisticated, appropriate and reliable manner.

  Further, according to the invention of Example 1, when a face image of an individual is continuously captured until a predetermined time has elapsed since the start of imaging, and the face image of the front of the individual is captured, When an identification check collation score is calculated using the feature amount of the face image, a plurality of front face images are captured, and a plurality of identification confirmation collation scores are calculated according to the plurality of front face images, The combination of the face image of the front of the person compared when calculating the highest ID verification score and the highest ID verification score, among the multiple ID verification scores Since one or more of the combinations of the face images of the front faces of the individuals compared in calculating the value identification verification score and the lowest value identification verification score are output to a predetermined output unit, for example, When capturing a person's face with a camera, etc., multiple biographs are taken consecutively to prevent imaging errors and the front face necessary for verification is detected. It is possible to execute the matching that has been performed at an advanced, appropriate and reliable level. In addition, when a large number of face images are captured and identity verification (identity confirmation) is performed, only a small amount of important verification results are extracted from the verification results and used for verification by the operator. Collation using information can be executed in an advanced, appropriate and efficient manner. Furthermore, for example, if the operator confirms only the best photograph (the face image with the highest matching score), the possibility of mistakenly identifying another person who is similar is the worst photograph. If the operator confirms only (the face image with the lowest matching score), there is a higher possibility that a dissimilar person will be mistakenly confirmed not to be the person, but which policy will be used for confirmation? Because it is output according to the policy according to the usage pattern, such as whether to check with a policy that combines both, or it is possible to perform collation using personal biometric information in a sophisticated, appropriate and operational format Become. In addition, since it is possible to quantitatively determine the appearance of the face image using the matching score with the face photo of an ID card such as a driver's license, it is possible to execute matching using personal biometric information in a sophisticated and appropriate manner. It becomes possible.

  By the way, although the personal biometric information collation apparatus according to the first embodiment has been described so far, the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, different embodiments will be described below as the personal biometric information matching device according to the second embodiment.

[Personal biometric information]
In the first embodiment, the case where the feature amount of the face image obtained by capturing the face is used as the feature amount of the personal biometric information is described. However, the present invention is not limited to this, and the feature amount of the personal biometric information is used. As described above, the present invention can be similarly applied to a case where a feature amount of personal biometric information that is difficult to forge completely is used, such as a case where a feature amount of a fingerprint or a feature amount of a vein is used.

[Imaging]
Further, in the first embodiment, the personal biometric information collating apparatus includes an identification image capturing unit (for example, a scanner) that captures an identity certificate issued by a public institution in which the personal biometric information is described. Although the case where the feature amount of the personal biometric information is calculated from the personal biometric information imaged by the identification image capturing means has been described, the present invention is not limited to this, and the personal biometric information collating device is used as the identification certificate image capturing means. The present invention can be similarly applied to the case where the feature amount of the personal biometric information is calculated from the personal biometric information acquired by other means.

  In the first embodiment, the personal biometric information collating apparatus includes personal imaging means (for example, a camera) that images personal biometric information of an individual who is a subject, and the personal biometric information captured by the personal imaging means. However, the present invention is not limited to this, and the personal biometric information matching device does not include the personal imaging unit, and the personal biometric information is obtained from the personal biometric information acquired by other means. The present invention can also be applied to the case where the feature amount is calculated.

[Comparison score calculation]
In the first embodiment, the database classifies and stores the feature quantity of the personal biometric information into the identification certificate feature quantity and the personal feature quantity, and the matching score calculation means captures the identification certificate image. Compare the feature quantity of the personal biometric information captured by the means with the identification-type feature quantity to calculate a matching score, and compare the feature quantity of the personal biometric information captured by the personal imaging means with the personal feature quantity However, the present invention is not limited to this, and the database classifies the feature values of personal biometric information into identification feature values and personal feature values. The present invention can also be applied to the case where the collation score calculation means calculates the collation score regardless of the classification of the identification system feature quantity and the personal system feature quantity.

  In the first embodiment, the personal biometric information matching device compares the feature amount of the personal biometric information captured by the identification image capturing unit with the feature amount of the personal biometric information captured by the personal image capturing unit. In the above description, the identification confirmation collation score indicating the degree of collation between the personal biometric information is calculated. However, the present invention is not limited to this, and the present invention is applicable to the case where the identification confirmation collation score is not calculated. Can be applied as well.

[output]
In the first embodiment, the personal biometric information collation apparatus associates the collation score with the data (face image) related to the individual compared when calculating the collation score as a result of the collation, and outputs a predetermined value. However, the present invention is not limited to this, and not only outputs the matching score and the data (face image) related to the individual compared in calculating the matching score in association with each other. As shown in FIG. 14, when the data (name) related to the individual is also output to the predetermined output unit, or when the matching score and the data (name) related to the individual are output in association with each other, Any method may be used as long as the result of collation is output to a predetermined output unit. As shown in FIG. 14, for example, when data (name) related to an individual is also output to a predetermined output unit, an operator who confirms the output is not limited to the face image collation score. It is possible to check the degree of similarity of names.

  In the first embodiment, the case where the personal biometric information collation apparatus arranges the collation results in the descending order of the calculated collation score values and outputs them to a predetermined output unit has been described. However, the order of output is not limited to this, and may be output in any order, for example, when output is performed in the order stored in the database.

  In the first embodiment, the personal biometric information collation apparatus outputs the collation result to a predetermined output unit in a form processed using a gauge that visually expresses the calculated collation score value. However, the present invention is not limited to this. For example, as shown in FIG. 15, the result of collation is output in a form processed using a meter that visually represents the collation score value. 16 or when the result of the collation is output in a form processed using a mark that visually expresses the value of the collation score as shown in FIG. 16, or as shown in FIG. When the result of matching is output in a form processed using a color that visually expresses the value, or in a form processed using a font that visually represents the value of the matching score as shown in FIG. In the case of outputting the collation result, FIG. As shown in FIG. 20, when the result of matching is output in a form processed using a size that visually expresses the value of the matching score, or as shown in FIG. In other cases, such as when outputting collation results in a form that is processed using a radar chart that combines the degree of collation of such information, or when outputting collation results in a form that is processed by combining a plurality of these, etc. The present invention can be similarly applied to the case where the collation result is output in the form.

  Further, in the first embodiment, the personal biometric information collating device is used when a plurality of front face images are captured by the personal imaging unit and a plurality of identification confirmation collation scores are calculated by the identification confirmation collation score calculating unit. The method for outputting the combination of the face image of the front of the individual compared in calculating the highest value identification verification score and the highest value identification verification verification score has been described, but the present invention is not limited to this. The method of outputting the combination of the face images of the front of the person compared when calculating the lowest ID verification score and the lowest ID verification score, or the average of the ID verification scores Or a method of outputting a predetermined display indicating that the identification confirmation collation score exceeds the predetermined threshold, or a combination thereof. Such as a method of output Te, it may be either.

  In the first embodiment, the case where the personal biometric information collating apparatus does not perform the process of registering the data used for collation in the database has been described. However, the present invention is not limited to this, Personal biometric information captured by the hand-drawn image capturing means, feature amounts calculated from the personal biometric information, personal biometric information captured by the personal image capturing means, feature amounts calculated from the personal biometric information, and identity confirmation collation scores The present invention can also be applied to the case where a process of associating one or more and registering them in the database is performed.

[System configuration, etc.]
In addition, among the processes described in the above embodiment, the process described as being automatically performed (for example, after the collation score is calculated in the identification confirmation collation score calculation unit 33a, the identification confirmation collation result output unit 33b All or part of the process (such as processing to output the result of collation) can be performed manually (for example, after the collation score is calculated, the command is input and then the collation result is output), or The process described as being performed manually (for example, duplication confirmation instruction by an operator or the like) is automatically performed by a known method (for example, the identification confirmation collation score exceeds a predetermined threshold) In such a case, the collation score calculation process in the duplication confirmation collation score calculation unit 34a may be started). In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-mentioned document and drawings can be arbitrarily changed unless otherwise specified.

  Each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated (for example, FIG. 2). In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. They can be integrated and configured (for example, when the database holding unit 25 is configured with different hardware). Further, all or any part of each processing function performed in each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware by wired logic.

  The personal biometric information collating method described in the above embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program can be distributed via a network such as the Internet. The program can also be executed by being recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, and a DVD and being read from the recording medium by the computer.

  As described above, the personal biometric information collation apparatus, the personal biometric information collation program, and the personal biometric information collation method according to the present invention provide data related to individuals in a database that stores data related to individuals classified by individual, It is useful for collating personal biometric information related to a person and outputting a collation result, and is particularly suitable for performing collation using personal biometric information in a sophisticated and appropriate manner.

It is a figure for demonstrating the outline | summary and characteristic of the personal biometric information collation apparatus which concern on Example 1. FIG. It is a block diagram which shows the structure of the personal biometric information collation apparatus which concerns on Example 1. FIG. It is a figure for demonstrating an output part. It is a figure for demonstrating an identification card imaging data holding part. It is a figure for demonstrating a personal imaging data holding part. It is a figure for demonstrating an identification confirmation collation score holding | maintenance part. It is a figure for demonstrating a duplication confirmation collation score holding | maintenance part. It is a figure for demonstrating a database holding part. It is a figure for demonstrating a personal imaging data processing part. 6 is a flowchart illustrating a processing procedure performed by the personal biometric information collating apparatus according to the first embodiment. It is a flowchart which shows the procedure of identification image pick-up data processing. It is a flowchart which shows the procedure of an identification confirmation collation score calculation process. It is a flowchart which shows the procedure of duplication confirmation collation score calculation processing. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part. It is a figure for demonstrating an output part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Personal biometric information collation apparatus 11 Input part 12 Identification card imaging part 13 Personal imaging part 14 Output part 15 Input / output control IF part 20 Storage part 21 Imaging data holding part 21a Identification card imaging data holding part 21b Personal imaging data holding part 22 feature quantity holding unit 22a identification certificate feature quantity holding unit 22b personal feature quantity holding unit 23 identification confirmation collation score holding unit 24 duplication confirmation collation score holding unit 25 database holding unit 30 control unit 31 imaging data processing unit 31a identification card imaging Data processing unit 31b Personal imaging data processing unit 32 Feature amount calculation unit 33 Identification confirmation unit 33a Identification confirmation collation score calculation unit 33b Identification confirmation collation result output unit 34 Duplication confirmation unit 34a Duplication confirmation collation score calculation unit 34b Duplication confirmation collation result output unit

Claims (15)

A personal biometric information collating apparatus that collates data related to an individual in a database that stores data related to the individual and stores the data, and personal biometric information related to a predetermined individual, and outputs a result of the verification ,
Database holding means for holding the database for storing the feature quantity of the personal biometric information of the individual as the data relating to the individual;
Feature amount calculating means for calculating a feature amount of the personal biometric information from the personal biometric information related to the predetermined individual;
The feature quantity of the personal biometric information calculated by the feature quantity calculation means is compared with the feature quantity of the personal biometric information stored in the database held by the database holding means, and the degree of matching between the personal biometric information is determined. A matching score calculating means for calculating a matching score to be shown;
The personal biometric information collating apparatus comprising: output means for outputting the result of the collation determined by the collation score calculated by the collation score calculating means to a predetermined output unit. The database holding unit holds the database that stores the feature amount of a face image obtained by capturing a face as the feature amount of the personal biometric information,
The feature amount calculating means calculates a feature amount of a face image as a feature amount of the personal biometric information,
2. The collation score calculation unit calculates a collation score by comparing the feature amount of the face image calculated by the feature amount calculation unit with the feature amount of the face image stored in the database. The personal biometric information collating device described in 1. It further comprises an identification card imaging means for imaging an identification card issued by a public institution in which personal biometric information is described,
3. The personal biometric information collating apparatus according to claim 1, wherein the feature quantity calculating unit calculates a feature quantity of the personal biometric information from the personal biometric information captured by the identification image capturing unit. . It further comprises personal imaging means for imaging personal biometric information of the individual who is the subject,
The personal biometric information according to any one of claims 1 to 3, wherein the feature quantity calculating means calculates a feature quantity of the personal biometric information from the personal biometric information imaged by the personal imaging means. Verification device. The database held by the database holding means is an individual whose subject is an identification-based feature quantity calculated from the personal biometric information described in the captured identification card, and the subject is the feature quantity of the personal biometric information. The personal biometric information is classified and stored in the personal feature amount calculated from the personal biometric information captured.
The collation score calculation means, when the feature quantity of the personal biometric information is calculated by the feature quantity calculation means from the personal biometric information imaged by the identification card imaging means, the feature quantity and the identification system A comparison score is calculated by comparing with a feature quantity, and when the feature quantity of the personal biometric information is calculated by the feature quantity calculation means from the personal biometric information captured by the personal imaging means, the feature quantity and the The personal biometric information collation apparatus according to claim 4, wherein the collation score is calculated by comparing with the personal feature amount.   Calculated by the feature amount calculating means from the feature amount of the personal biometric information calculated by the feature amount calculating means from the personal biometric information imaged by the identification image capturing means and the personal biometric information imaged by the personal imaging means. 5. An identification confirmation collation score calculating means for comparing the characteristic amount of the personal biometric information thus obtained and calculating an identification confirmation collation score indicating a degree of collation between the personal biometric information. The personal biometric information collating device described in 1.   The output means associates the collation score calculated by the collation score calculation means with the data related to the individual compared when calculating the collation score as a result of the collation, and outputs the data to the predetermined output unit The personal biometric information collation apparatus according to any one of claims 1 to 6,   The output means arranges the results of the collation in descending order of the collation score values calculated by the collation score calculation means, and outputs the results to the predetermined output unit. One personal biometric information collation apparatus as described in one.   The output means is processed using any one or more of a gauge, a meter, a mark, a color, a font, and a size that visually represents the value of the collation score calculated by the collation score calculation means. The personal biometric information collation apparatus according to claim 1, wherein the collation result is output to the predetermined output unit.   The output means, when the identity confirmation collation score calculated by the identity confirmation collation score calculation means exceeds a predetermined threshold, displays a predetermined display indicating that the predetermined threshold exceeds the predetermined threshold. The personal biometric information collating device according to claim 6, wherein the personal biometric information collating device is output. The personal imaging means continuously captures the individual's face image until a predetermined time has elapsed since the start of imaging,
The identity confirmation collation score calculating means calculates the identity confirmation collation score using the feature quantity of the front face image when the face image of the front face of the individual is imaged by the personal imaging means,
The output means captures a plurality of front face images by the personal imaging means, and calculates a plurality of identity confirmation collation scores according to the plurality of front face images by the identity confirmation collation score calculation means. A combination of the face image of the front of the individual compared in calculating the highest value identification check score and the highest value confirmation check score among the plurality of identity check match scores, The predetermined output unit outputs one or more of a combination of the face image of the front of the individual compared in the calculation of the lowest value identification confirmation matching score and the lowest value identification confirmation matching score among the matching scores. The personal biometric information collating device according to claim 6, wherein the personal biometric information collating device is output. The personal imaging means continuously captures the individual's face image until a predetermined time has elapsed since the start of imaging,
The identity confirmation collation score calculating means captures a plurality of identity confirmation collation scores when a plurality of the individual face images are imaged by the individual imaging means and a plurality of identity confirmation collation scores are calculated according to the plurality of face images. Calculate the average verification verification score,
The personal biometric information collation according to claim 6, wherein the output unit outputs the average value to the predetermined output unit when the average value is calculated by the identification confirmation collation score calculation unit. apparatus.   Personal biometric information captured by the identification image capturing means, feature amounts calculated by the feature amount calculating means from the personal biometric information, personal biometric information captured by the personal image capturing means, and the characteristics from the personal biometric information Either one or more of the feature quantity calculated by the quantity calculating means and the identity check matching score calculated by comparing the two feature quantities by the identity check matching score calculating means are registered in the database. 7. The personal biometric information collating apparatus according to claim 6, further comprising registration means. A personal living body that causes a computer to execute a method of collating data related to an individual in a database that stores data relating to the individual and storing the data, and personal biometric information relating to a predetermined individual, and outputting the result of the collation An information verification program,
A database holding procedure for holding the database for storing the feature quantity of the personal biometric information of the individual as the data relating to the individual;
A feature amount calculating procedure for calculating a feature amount of the personal biometric information from the personal biometric information related to the predetermined individual;
The feature quantity of the personal biometric information calculated by the feature quantity calculation procedure is compared with the feature quantity of the personal biometric information stored in the database held by the database holding procedure, and the degree of matching between the personal biometric information is determined. A matching score calculation procedure for calculating a matching score to be shown;
A personal biometric information collation program that causes a computer to execute an output procedure for outputting the collation result determined by the collation score calculated by the collation score calculation procedure to a predetermined output unit. A personal biometric information collation method that collates data related to an individual in a database that stores data relating to the individual and stores the data, and personal biometric information related to a predetermined individual, and outputs the result of the collation ,
A database holding step of holding the database for storing the feature quantity of the personal biometric information of the individual as the data relating to the individual;
A feature amount calculating step of calculating a feature amount of the personal biometric information from the personal biometric information related to the predetermined individual;
The feature quantity of the personal biometric information calculated by the feature quantity calculation step is compared with the feature quantity of the personal biometric information stored in the database held by the database holding step, and the degree of matching between the personal biometric information is determined. A matching score calculation step for calculating a matching score to be shown;
An output step of outputting the result of the collation determined by the collation score calculated by the collation score calculation step to a predetermined output unit.

Images