JP5564414B2 - Biometric authentication system and biometric authentication method - Google Patents

Description

  The present invention relates to a biometric authentication system and a biometric authentication method in a personal authentication method.

  Authentication processing using biometric information is performed by comparing biometric information acquired from a user with biometric information registered in advance, and there are 1: 1 authentication and 1: N authentication as authentication methods. . First, in 1: 1 authentication, biometric information to be collated is extracted from a database based on an identification number such as an ID card or key input that a user inputs into an authentication system. And the collation process with the biometric information extracted from the biometric information acquired from the user is performed. On the other hand, in 1: N authentication, based on the biometric information acquired from the user, verification with biometric information included in the database stored in the authentication system is repeated in order to find the same one.

  Using such an authentication method, for example, in Patent Document 1, identification information is obtained by executing 1: N authentication based on biometric information acquired from a user using a database in which users are grouped. It is disclosed that authentication processing is performed by specifying and executing 1: 1 authentication on all user databases using the identification information and the acquired biometric information.

JP 2004-145608 A

  However, when 1: N authentication is performed on all registrants, even registrants who cannot perform accurate 1: N authentication processing due to the low quality of registered biometric information must always perform 1: N authentication processing. The processing time is always long. In addition, since accurate 1: N authentication processing cannot be performed, there is a high possibility that another person will be accepted.

  Therefore, according to the present invention, a 1: 1 authentication user can start 1: 1 authentication without performing 1: N authentication, and a 1: N authentication user starts 1: N authentication. An object of the present invention is to provide a biometric authentication system that can perform such a process.

A biometric authentication system according to an embodiment of the present invention includes a biometric information input unit that inputs biometric information of a user, an identification information input unit that inputs identification information of the user, and a verification process using the biometric information of the user A biometric authentication verification unit that performs the control, a control unit that controls the biometric authentication verification unit based on an input from the biometric information input unit or the identification information input unit, and a biometric information that has been given identification information and a biometric that has not been given identification information And a biometric information stored in the collation database when the biometric information input unit acquires biometric information without the identification information input unit acquiring identification information. the door against the biometric authentication checking unit, the biometric information input unit acquires biometric information, and when the identification information input unit acquires the identification information, stored in the verification database corresponding to the acquired identification information And controlling to determine authentication success or failure to collate the acquired biometric information and body information.

  According to the present invention, it is possible to provide a biometric authentication system and a biometric authentication method having a short authentication processing time while maintaining an authentication success rate.

A biometric authentication system according to an embodiment of the present invention Detailed configuration of a biometric authentication system according to an embodiment of the present invention Biometric information registration processing according to an embodiment of the present invention Biometric information quality discrimination processing according to an embodiment of the present invention 1: 1 collation processing according to an embodiment of the present invention An example of 1: N verification processing according to an embodiment of the present invention Biometric authentication processing according to Embodiment 1 of the present invention Registration processing to division collation DB concerning Example 2 of the present invention The biometric authentication process at the time of using the division | segmentation collation DB concerning Example 2 of this invention. Authentication processing using barcode reading processing according to Embodiment 3 of the present invention Authentication processing in division verification DB using barcode reading processing according to Embodiment 4 of the present invention Biometric authentication processing according to embodiment 5 of the present invention Biometric authentication processing according to embodiment 6 of the present invention Application example of collation DB in biometric authentication system according to an embodiment of the present invention

  While 1: 1 authentication can perform relatively high-precision personal authentication, the user needs to carry an ID card or a bar code or memorize identification information to be keyed, which is not convenient. In 1: N authentication, the user only needs to present biometric information and is highly convenient, but the time required for verification is longer than in 1: 1 authentication. Further, there is a higher possibility of accepting another person that the matching with the feature amount created from the other person's biometric information is erroneously successful compared to the 1: 1 authentication.

  Therefore, the present invention has the following characteristics.

First, a biometric authentication system according to an embodiment of the present invention uses a biometric information input unit that inputs biometric information of a user, an identification information input unit that inputs identification information of the user, and biometric information of the user. A biometric authentication verification unit that performs verification processing, a control unit that controls the biometric authentication verification unit based on an input from the biometric information input unit or the identification information input unit, and biometric information and identification information to which identification information is added And when the biometric information input unit acquires biometric information without the identification information input unit acquiring the identification information, the biometric information stored in the verification database is stored. a biometric information verification biometric authentication verification unit, the biometric information input unit acquires biometric information, and when the identification information input unit acquires the identification information, stored in the verification database corresponding to the acquired identification information And determining a control to authenticate the success or failure to collate the biometric information and the acquired physiological information. More specifically, when the biometric information input unit acquires biometric information without the identification information input unit acquiring the identification information, an unmatched biometric information feature amount is acquired from the verification database and acquired. When the biometric information feature amount of the biometric information is collated, the biometric information input unit acquires the biometric information, and the identification information input unit acquires the identification information, the biometric information feature amount corresponding to the acquired identification information is obtained from the collation database. It is acquired and collated with the biometric information feature amount of the acquired biometric information. Thus, a user having high-quality biometric information can perform a reliable authentication process by presenting a registered biometric information, and a user having low-quality biometric information inputs pre-assigned identification information. By doing so, it is possible to perform a quick and reliable authentication process. Here, the identification information is information that is issued from the authentication system when the user registers the biometric information in the authentication system, and is information that is allocated so as not to overlap for each user or biometric information.

  Next, in the biometric authentication system according to the embodiment of the present invention, the collation database is lower in quality than the first collation database storing a plurality of pieces of biometric information and the biometric information stored in the first collation database. The biometric authentication collating unit is divided into a biometric information collating unit in which a plurality of biometric information is stored in association with identification information. When the biometric information of the user is acquired, the acquired biometric information is collated with the biometric information stored in the first collation database, the biometric information input unit acquires the biometric information of the user, and is identified. When the information input unit acquires the identification information of the user, the biological information stored in the second verification database corresponding to the acquired identification information is compared with the acquired biological information. More specifically, when the biometric information input unit does not acquire the identification information and the biometric information input unit acquires the biometric information, the unmatched biometric information feature amount is acquired from the first verification database and acquired. When the biometric information is compared with the biometric feature quantity of the biometric information, the biometric information input unit acquires the biometric information, and the identification information input unit acquires the identification information. It is obtained from a collation database and collated with the biometric information feature amount of the obtained biometric information. As a result, a user having high-quality biometric information can preferentially collate with a collation database in which high-quality biometric information is stored by presenting the registered biometric information. Can be achieved. In addition, for users who have low quality biometric information, by inputting pre-assigned identification information, it is possible to preferentially collate with a verification database in which low quality biometric information is stored. Rapid authentication processing can be performed.

Furthermore, in the biometric authentication system according to the embodiment of the present invention, the control unit includes an input information determination unit that determines whether identification information and biometric information are input, and the input information determination unit acquires the identification information. In this case, when the biometric authentication verification unit compares the acquired biometric information with the biometric information stored in the verification database, and the input information determination unit does not acquire the identification information, it corresponds to the acquired identification information. The biometric authentication verification unit controls the biometric information stored in the verification database to be compared with the acquired biometric information.

  Further, in the biometric authentication system according to the embodiment of the present invention, the biometric information input unit has a biometric detection unit that detects a biometric, and when the biometric detection unit detects the biometric, the control unit inputs biometric information. When the living body information is read by the unit and the living body is not detected, the barcode reading process is operated. As described above, even when barcode information is used as identification information, it is possible to achieve downsizing of the individual authentication system by adopting a configuration in which the biometric information input unit also serves as the identification information input unit.

  Furthermore, the biometric authentication system according to the embodiment of the present invention is the case where the biometric information input unit acquires the biometric information of the user without the identification information input unit acquiring the biometric information of the user. When collation fails as a result of collation with the biometric information stored in the first collation database, the user is prompted to input identification information, and based on the inputted identification information, the first collation database or the first collation information Biometric information is extracted from the collation database 2 and collation processing is performed. Thus, even if the collation fails due to improper placement of the living body or when the user who should perform the 1: 1 authentication by inputting the identification information incorrectly presents the living body, 1 again. : Authentication processing can be performed by performing 1: 1 authentication with a verification database for N authentication or 1: 1 authentication.

  In addition, the biometric authentication system according to the embodiment of the present invention is the case where the biometric information input unit acquires the biometric information of the user without the identification information input unit acquiring the biometric information of the user, When collation fails as a result of collation with the biometric information stored in the first collation database, the biometric information stored in the second collation database is re-collated using the acquired biometric information. When the verification is successful, the user is prompted to input identification information, and the input identification information is compared with the identification information corresponding to the biometric information successfully verified. As described above, when a user who should perform 1: 1 authentication by inputting identification information erroneously presents a biometric, the biometric information is used to perform 1: N authentication in the verification database for 1: 1 authentication. In addition to processing, the configuration in which identification information is input and the identification information is collated makes it possible to minimize the acceptance of others in the authentication process for a user with low quality biometric information.

Next, a biometric authentication system according to an embodiment of the present invention includes a biometric information registration terminal, and the biometric information registration terminal includes a biometric information acquisition unit that acquires biometric information from a user, and biometric information from the acquired biometric information. A control unit that extracts a feature quantity; and a quality determination unit that judges the quality of the extracted biometric information feature quantity. When the quality of the biometric information feature quantity is equal to or higher than a specified value, the control section Is stored in the first collation database, and when the quality of the biometric information feature quantity is less than the specified value, the biometric information feature quantity is stored in the second collation database in association with the identification information of the user. And In addition, the quality determination unit determines the quality based on the number of pixels identified as a blood vessel in the acquired biological information.
Example 1
Hereinafter, embodiments according to the present invention will be described in detail with reference to the drawings. First, FIG. 1 is a diagram showing a configuration of a biometric authentication system according to an embodiment of the present invention. As shown in FIG. 1, a biometric authentication system 100 includes a biometric information input unit 101 that acquires biometric information of a user, a biometric authentication verification unit 102 that performs a verification process based on the acquired biometric information, and a biometric of the user. A quality determination unit 103 that determines the quality of information, a display unit 104, a control unit 105 that controls the above-described components, an identification information issuing unit 106 that issues identification information to the user, and an identification issued to the user An information input unit 107, an identification information management unit 108, and a collation DB 109 are provided. In this biometric authentication system 100, the biometric authentication collation unit 102 collates the feature amount data extracted from the biometric information input from the biometric information input unit 101 with the feature amount data stored in the collation DB 109, thereby enabling the user. Identity verification can be implemented. The biometric authentication verification unit 102 performs authentication processing according to the quality of the feature quantity for each user determined by the quality determination unit 103. When performing 1: 1 authentication, identification information issued by the identification information issuing unit 106 at the time of biometric information registration is used. At the time of authentication, the identification information management unit 108 identifies whether the identification information input from the user by the identification information input unit 107 is correct. If the identification information is correct, the feature corresponding to the identification information Quantity data is extracted from the collation DB 109, and the biometric authentication collation unit 102 performs 1: 1 collation processing using the feature quantity data and the feature quantity generated from the biometric information input from the biometric information input unit 101. The identification information input unit 107 may be a numeric keypad, a barcode, a contact card reader, a non-contact card reader, and other input methods. The embodiment of the identification information input unit 107 is not specified.

  When implementing 1: N authentication, the feature-value produced | generated from the biometric information input from the biometric information input part 101 is collated with each feature-value data stored in collation DB109. The verification DB 109 stores all the 1: 1 authentication biometric information data and 1: N authentication biometric information data, but the 1: 1 authentication biometric information data and the 1: N authentication biometric information data. It is more preferable to separate and separately store them in the verification DB. The control unit 105 controls the exchange of information between the functional blocks.

  In the following description, the description is made on the assumption that the display unit 104 prompts the user to present biometric information and the authentication result is displayed as visual information. It may be performed by means other than sight, or nothing may be transmitted.

  Next, a more detailed biometric authentication system will be described with reference to FIG. In the example illustrated in FIG. 2, a biometric information registration terminal 202, a biometric information authentication terminal 203, and a verification DB 109 are connected via a network 201. The biometric information registration terminal 202 includes a computer 204, a biometric information acquisition device 205, an identification information issuing unit 206, and a printer 207. The computer 204 includes a biometric information registration terminal control unit 212, a quality determination unit 213, and the like. The biometric information acquisition device 205 includes a biometric information acquisition device control unit 210 and a biometric information input unit 211. The display unit 214 includes an identification information management unit 215. The biometric information authentication terminal 203 includes a computer 208, a biometric authentication device 209, and an identification information input unit 222. The computer 208 includes a biometric authentication terminal control unit 216, a display unit 217, and an identification information management unit 218. The biometric authentication device 209 includes a biometric authentication device control unit 219, a biometric information input unit 220, and a biometric authentication verification unit 221.

  In the configuration shown in FIG. 2, when registering biometric information, the user presents biometric information to the biometric information input unit 211 of the biometric information acquisition device 205 of the biometric information registration terminal 202. The feature amount data generated from the presented biometric information is stored in the collation DB 109, and when the quality determination unit 213 determines that the quality of the feature amount data is not high, the identification information issue unit 206 identifies the feature amount data. Information is issued. The issued identification information is printed on a barcode by the printer 207, and the barcode printed by the user is held. The barcode used here may be a one-dimensional barcode, a two-dimensional barcode, or a barcode according to another embodiment.

Authentication processing using the registered biometric information is executed by the biometric information authentication terminal 203. When the user has a barcode, the user inputs the barcode from the identification information input unit 222 and presents the biological body to the biological information input unit 220 of the biometric authentication device 209. The biometric authentication terminal control unit 216 inquires the identification information management unit 218 about the identification information read from the barcode, and if the identification information is correct, the biometric authentication terminal control unit 216 collates the feature data for matching corresponding to the identification information. Obtained from the DB 109 and notified to the biometric authentication device control unit 219, the biometric authentication collation unit 221 performs 1: 1 collation between the acquired feature data for collation and feature data extracted from the biometric information. . When the user does not have a barcode, the user presents the biometric to the biometric information input unit 220 of the biometric authentication device 209. The biometric authentication collation unit 221 uses the feature amount data extracted from the biometric information, and performs 1: N collation with the feature amount data for collation in the collation DB 109. The verification result is displayed on the display unit 217.
The configuration according to the present invention is not limited to the above-described embodiment, and the verification DB 109 is built in the biometric information authentication terminal 203, the biometric authentication verification unit 221 is built in the computer 208, or the biometric information registration process. Various configurations such as performing biometric information authentication processing with a single terminal are conceivable.

  FIG. 3 is a processing flowchart when registering feature data for verification in the database for verification in the biometric authentication system according to the present embodiment. First, user information is input by the user, and user information necessary for the system, such as the user's name, is registered (step 301). Thereafter, the user presents the living body to the biometric information input unit and obtains the biological information (step 302). Next, feature data for verification is extracted from the acquired biometric information (step 303). It is determined whether or not the extracted feature data is high quality (step 304). If it is determined that the extracted feature data is high quality, the feature data is registered in the verification DB (step 305). If it is determined that the extracted feature amount data is not of high quality in the determination processing of the feature amount data, identification information is issued to the user (step 306). Then, the identification information and the feature amount data are registered in the collation DB (step 305). The identification information issued here is for use in 1: 1 authentication processing. Further, high quality feature amount data means that there is a lot of information for specifying an individual in the feature amount data. On the other hand, low quality feature data means that there is little information for identifying individuals in the feature data, and as a result, there is a high possibility that matching with another individual will be erroneously successful. It means to become.

  In FIG. 3, the user presents the living body and extracts the feature amount data for collation one time each. However, in order to acquire more stable biological information and feature amount, each time, multiple times are obtained. You may carry out. Further, in this embodiment, the case where finger vein information is used as the biometric information is mentioned. However, the present invention is not limited to the finger vein information, and other biometric information such as a fingerprint, a palm, and an iris may be used. I do not care.

  FIG. 4 shows a flowchart of the quality determination process. First, biometric information is obtained from the biometric presented by the user, and image data is obtained (step 401). Next, feature data is extracted from the image data (step 402). Then, it is determined whether or not the number of pixels identified as a blood vessel is equal to or greater than a predetermined value. Specifically, for each pixel included in the feature amount data, whether the blood vessel is or is not identified based on the color and brightness is determined. If the number of pixels identified as is greater than or equal to a predetermined value, it is determined that the quality is high, and if the number of pixels is less than the predetermined value, it is determined that the quality is low (step 403). Here, the prescribed value that serves as a criterion for discrimination is a value determined for each biometric authentication system.

  Note that FIG. 4 is based on a system that authenticates the identity of a blood vessel in a part of a user's body as biometric information, but the embodiment is not limited to the case where the blood vessel distribution status is used as biometric information. . 4 is based on the premise that the feature amount data is handled as image data and authentication is performed, but the feature point extracted from the biometric information is handled and the authentication is performed in the system that performs authentication. A configuration is also conceivable in which the quality is discriminated from the mutual relationship of the feature points.

  FIG. 5 is a flowchart of the 1: 1 verification processing 500 when identification information is input and a living body is presented by the user. The 1: 1 verification process is started when the identification information and the biological information are acquired from the user. Based on the identification information input by the user, feature data for collation is acquired from the collation DB (step 501). Subsequently, the feature amount data extracted from the biometric information is compared with the feature amount data extracted from the collation DB (steps 503 and 504). Here, if it is determined that they are the same, the collation is successful (step 505), and if they are not the same, the collation is failed (step 506).

  In FIG. 5, after acquiring both identification information and biometric information from the user, feature quantity data corresponding to the identification information is acquired (step 501). First, only the identification information is acquired and then the identification is performed. The feature amount data corresponding to the information may be acquired, and then the biometric information presented by the user may be acquired, and then the feature amount data may be compared (step 503).

  FIG. 6 is a flowchart of 1: N collation processing 600 when a living body is presented by a user. The 1: N verification process is started when biometric information is acquired without identification information from a user. When the 1: N matching process is started, one unmatched feature data for matching is extracted from the matching DB 602 (step 601). Subsequently, the feature amount data extracted from the biometric information is compared with the feature amount data for matching extracted from the matching DB (steps 603 and 604). Here, if it is determined that they are the same, the collation is successful (step 605). If it is determined that they are not the same, whether or not all the unmatched feature data stored in the collation DB have been compared. If the determination is made and the determination is made, the verification fails (step 607).

  On the other hand, if not done for all, the process returns to step 601 to extract unmatched feature data and repeat the processing from step 601 onward. In FIG. 6, when data that succeeds in collation is found in the collation DB 602, the collation success 805 is output at that point and the process ends. It is good also as a structure which always performs comparison judgment with. In this case, when the comparison with all data is completed, if there is only one data that has been successfully verified, the authentication is successful. However, if the verification is successful with a plurality of data, the authentication may be unsuccessful. A method may be used in which identification information is input or a user's name is displayed for a plurality of successful candidates to be selected by the user.

  Next, FIG. 7 shows a processing flow of the biometric authentication system according to the first embodiment.

  FIG. 7 is a flowchart when the 1: 1 authentication process and the 1: N authentication process are used together in the biometric authentication system 100 shown in FIG. First, it waits for input of identification information or a living body (step 701). Whether or not the identification information is input is determined by the input information determination unit (step 702). When the input of the identification information is recognized, the biometric information is acquired (step 703), and the 1: 1 verification process is performed. This is performed by the authentication verification unit (step 704). Thereafter, it is determined whether or not the verification is successful (step 705). If the verification is successful, the authentication is successful (step 710), and if the verification fails, the authentication is failed (step 709). If the input of identification information is not recognized in step 702, biometric information is acquired (step 706), and 1: N matching processing is executed (step 707). Thereafter, it is determined whether or not the collation is successful (step 708). If the collation is successful, the authentication is successful (step 710), and if the collation is unsuccessful, the authentication is unsuccessful (step 711).

In FIG. 7, authentication failure 709 and 711 are performed when the verification is not successful in steps 705 and 708, but biometric information is acquired when the verification is not successful in steps 705 and 708. A system that returns to steps 703 and 706 and tries the collation process a plurality of times is also conceivable.
(Example 2)
As described in the first embodiment, since the feature value data of all users is stored in the matching DB 109, feature amount data with low quality in the 1: N matching process is also a matching target. Therefore, a second embodiment in which the collation DB 109 described in FIGS. 1 and 2 is divided into a collation DB 810 for 1: N authentication and a collation DB 811 for 1: 1 authentication will be described with reference to FIGS. To do. First, the registration process will be described. As in FIG. 3, user information is input by the user, and user information necessary for the system, such as the user's name, is registered (step 801). Thereafter, the user presents the living body to the living body information input unit and obtains the biological information (step 802). Next, feature data for verification is extracted from the acquired biometric information (step 803). It is determined whether or not the extracted feature data is high quality (step 804). If it is determined that the quality is high, the feature data is registered in the 1: N authentication verification DB (step). 805). When it is determined that the quality of the extracted feature data is not high, identification information is issued to the user (step 806). Then, the identification information and the feature amount data are registered in the 1: 1 authentication verification DB (step 807). As described above, high quality of feature data means that there is a lot of information for identifying individuals in the feature data, while low quality of feature data means that This means that there is little information for specifying an individual, and as a result, there is a high possibility that collation with another individual will be erroneously successful.

  Further, the quality determination process is the same as the method described in FIG. 4, and as in FIG. 3, the presentation of the biometric by the user and the extraction of the feature amount data for verification are more stable. Each may be performed multiple times to obtain the quantity. The biometric authentication system in FIG. 9 according to the second embodiment first waits for input of either identification information or a biometric (step 901). The input information determination unit determines whether or not the identification information is input (step 902). When the input of the identification information is recognized, the biometric information is acquired (step 903), and the 1: 1 authentication verification DB 811 is used. The 1: 1 verification process is performed by the biometric authentication verification unit (step 904). Thereafter, it is determined whether or not the verification is successful (step 905). If the verification is successful, the authentication is successful (step 910), and if the verification fails, the authentication is unsuccessful (step 909). If the input of identification information is not recognized in step 902, biometric information is acquired (step 906), and 1: N verification processing in the 1: N authentication verification DB 810 is executed (step 907). . Thereafter, it is determined whether or not the verification is successful (step 908). If the verification is successful, the authentication is successful (step 910), and if the verification is unsuccessful, the authentication is failed (step 911).

In FIG. 9, authentication failure 909 and 911 are determined if the verification is not successful in steps 905 and 908, but biometric information is acquired if the verification is not successful in steps 905 and 908. A system that returns to steps 903 and 906 and tries the collation process multiple times is also conceivable. By adopting such a configuration, high-quality feature data is stored in the 1: N authentication verification DB 810 to be used. Therefore, only high-quality feature data is used in the 1: N verification processing in FIG. Since the matching process is carried out using this, the possibility of erroneous matching in the 1: N matching process can be further reduced as compared with FIG. In addition, the processing time can be shortened. In addition, about the structure and process which are not mentioned in Example 2, it is the same as the structure and process of Example 1.
(Example 3)
In the first and second embodiments, the type of biometric information and the acquisition method are not specified, and the type of identification information and the input method are not specified. Here, in a biometric authentication system that satisfies a predetermined condition, the biometric information input unit and the identification information number input unit may be able to be shared. For example, it is a case of a biometric authentication system that uses biometric information acquired by capturing an image with a camera and acquires identification information by capturing an image with a camera. Hereinafter, it demonstrates in detail using FIG.

  FIG. 10 is a flowchart relating to a biometric authentication system equipped with a camera that also serves as a biometric information input unit and an identification information input unit. First, it is detected that something is held over the camera corresponding to the biological information input unit (identification information input unit) (step 1001). This detection process may be performed by continuously capturing images with a camera and detecting a difference in image from when nothing is held over, or may include a sensor dedicated to detection. If it is detected in step 1001 that something is held over, it is determined whether or not it is a living body (step 1002). If it is determined that the information is biometric information, biometric information is acquired (step 1003), and 1: N matching processing is executed (step 1004). Thereafter, by determining whether or not the verification is successful (step 1005), if the verification is successful, the authentication is successful (step 1013), and if the verification fails, the authentication is failed (step 1014).

On the other hand, if the living body is not detected in step 1002, barcode detection is executed (step 1006). Here, when the barcode is not detected, it is determined that the barcode is not correctly placed, and the barcode is detected again after instructing the user to place the barcode again (steps 1007 and 1006). ). When a barcode is detected in step 1006, identification information is acquired by barcode reading processing (step 1008), biometric information is acquired (step 1009), and 1: 1 verification processing is executed (step 1010). . Thereafter, by determining whether or not the collation is successful (step 1011), if the collation is recognized as successful, the authentication is successful (step 1013), and if the collation is not recognized as successful, the authentication is failed (step 1014). ). In addition, when it is not recognized that the collation is successful in steps 1005 and 1011, the biometric information acquisition 1003 and 1019 may be returned to try a plurality of 1: N collation processes or 1: 1 collation processes. In addition, about the structure and process which are not mentioned in Example 3, it is the same as the structure and process of Example 1.
(Example 4)
As in the third embodiment, a biometric authentication system that uses biometric information acquired by capturing an image with a camera and acquires identification information by capturing an image with a camera will be described with reference to FIG. This will be described in detail. FIG. 11 is a flowchart relating to an authentication system equipped with a camera serving as both a biometric information input unit and an identification information input unit. First, it is detected that something has been held over the camera corresponding to the biological information input unit (identification information input unit) (step 1101). This detection process may be performed by continuously capturing images with a camera and detecting a difference in image from when nothing is held over, or may include a sensor dedicated to detection. If it is detected in step 1101 that something is held over, it is determined whether or not it is a living body (step 1102). If it is determined that the information is biometric information, biometric information is acquired (step 1103), and 1: N verification processing is executed in the 1: N authentication verification DB (step 1104). Thereafter, by determining whether or not the collation is successful (step 1105), if the collation is successful, the authentication is successful (step 1113), and if the collation is unsuccessful, the authentication is unsuccessful (step 1114).

  On the other hand, if the living body is not detected in step 1102, barcode detection is executed (step 1106). Here, if the barcode is not detected, it is determined that the barcode is not correctly placed, and the barcode is detected again after instructing the user to place the barcode again (steps 1107 and 1106). ). When a bar code is detected in step 1106, identification information is acquired by a bar code reading process (step 1108), and biometric information is acquired (step 1109). Processing is executed (step 1110). Thereafter, by determining whether or not the collation is successful (step 1111), if the collation is recognized as successful, the authentication is successful (step 1113), and if the collation is not recognized as successful, the authentication fails (step 1114). ).

In addition, when it is not recognized that the collation is successful in Steps 1105 and 1111, the biometric information acquisition 1103 and 1119 may be returned, and a plurality of 1: N collation processes or 1: 1 collation processes may be attempted. In addition, about the structure and process which are not mentioned in Example 4, it is the same as the structure and process of Example 1. FIG.
(Example 5)
FIG. 12 according to the fifth embodiment is an embodiment in the case where the verification DB is divided into a 1: N authentication verification DB and a 1: 1 authentication DB according to the quality of biometric information. In the fifth embodiment, it is assumed that identification information is issued to all users regardless of the quality of biometric information. First, it waits for input of identification information or a living body (step 1201). It is determined whether or not the identification information has been input (step 1202). If the identification information has been input, biometric information is acquired (step 1203), and the 1: 1 authentication collation DB 811 has a 1: 1 ratio. The verification process is performed by the biometric authentication verification unit (step 1204). Thereafter, it is determined whether or not the collation is successful (step 1205). If the collation is successful, the authentication is successful (step 1214), and if the collation is unsuccessful, the authentication is unsuccessful (step 1213).

  If the input of identification information is not accepted in step 1302, biometric information is acquired (step 1206), and 1: N verification processing in the 1: N authentication verification DB 810 is executed (step 1207). . Thereafter, it is determined whether or not the collation is successful (step 1208). If the collation is successful, the authentication is successful (step 1214). If the collation is unsuccessful, input of identification information assigned in advance to the user is requested ( Step 1209). Then, after acquiring the identification information, a 1: 1 verification process is performed between the feature amount data corresponding to the identification information in the 1: N authentication verification DB and the feature amount data acquired in step 1206 (step 1211). . As a result, if the verification is successful, the authentication is successful (step 1214). On the other hand, if the verification fails, the authentication fails (step 1215).

  Here, after the 1: N verification process in the 1: N authentication verification DB, the 1: 1 verification process is performed again in the same 1: N authentication verification DB. In some cases, it is possible to authenticate again even when collation fails, and the convenience of the user is improved.

The configuration and processing not mentioned in the fifth embodiment are the same as the configuration and processing of the first embodiment.
(Example 6)
FIG. 13 according to the sixth embodiment is an embodiment in the case where the collation DB is divided into the 1: N authentication collation DB and the 1: 1 authentication DB according to the quality of the biometric information, as in the fifth embodiment. is there. In the sixth embodiment, first, input of either identification information or a living body is awaited (step 1301). It is determined whether or not the identification information is input by the input information determination unit (step 1302). When the input of the identification information is recognized, the biometric information is acquired (step 1303), and the 1: 1 authentication verification DB 811 is used. The 1: 1 verification process is performed by the biometric verification unit (step 1304). Thereafter, it is determined whether or not the verification is successful (step 1305). If the verification is successful, the authentication is successful (step 1316), and if the verification fails, the authentication is failed (step 1315).

  If the input of identification information is not accepted in step 1302, biometric information is acquired (step 1306), and 1: N verification processing in the 1: N authentication verification DB 810 is executed (step 1307). . Thereafter, it is determined whether or not the collation is successful (step 1308). If the collation is successful, the authentication is successful (step 1316). If the collation is unsuccessful, the biometric information is used in the 1: 1 authentication DB. The 1: N matching process is executed (step 1310). Here, if collation fails, authentication fails (step 1317). On the other hand, if the collation is successful, the user is requested to input identification information, and the identification information is compared with the identification information corresponding to the feature amount data that has been collated successfully to determine whether they match (step 1314). . As a result, if the collation is successful, the authentication is successful (step 1316). On the other hand, if collation fails, authentication will fail (step 1317). In this embodiment, there is a possibility that some users may present biometric information without inputting identification information even though the 1: 1 authentication process should be originally executed. In 1308, it is regarded as a collation failure. At this time, since the feature amount data of the presented biometric information should be included in the 1: 1 authentication verification DB 811, the feature amount to be verified by performing 1: N verification processing using the 1: 1 authentication verification DB 811. A 1: 1 authentication process is finally performed by searching for data and confirming the matching between the corresponding identification information and the input identification information (steps 1313 and 1314).

  In step 1311, the user is requested to input identification information by screen display.

  Further, when inputting identification information in step 1312 using the keyboard, if it is a burden for the user to store the identification information, in step 1311, instead of displaying a screen for input of identification information, verification is performed. The name of the successful user may be displayed, the user may determine whether or not it matches the user's name, and the user may input the determination result in step 1312. In addition, about the structure and process which are not mentioned in Example 6, it is the same as the structure and process of Example 1.

  Here, the verification DB will be described with reference to FIG. In the example illustrated in FIG. 14, identification information, a feature amount, and a user ID are stored in the verification DB. In the verification DB, feature quantities are managed by IDs that do not overlap. On the other hand, in the account DB, user ID, branch information, and account information are stored so that it can be understood which user has which account of which branch. In the example of FIG. 14, the account information of the user authenticated by biometric authentication can be acquired by linking the verification DB and the account DB using the user ID.

DESCRIPTION OF SYMBOLS 100 ... Biometric authentication system, 101 ... Biometric information input part, 102 ... Biometrics authentication collation part, 103 ... Quality determination part, 104 ... Display part, 105 ... Control part, 106 ... Identification information issuing part, 107 ... Identification information input part, 108 ... identification information management unit, 109 ... collation DB, 810 ... 1: N authentication collation DB, 811 ... 1: 1 authentication collation DB.

Claims (12)

A biometric information input unit for inputting biometric information of the user;
An identification information input unit for inputting user identification information;
A biometric authentication verification unit that performs verification processing using the biometric information of the user;
A control unit that controls the biometric verification unit based on an input from the biometric information input unit or the identification information input unit;
A collation database storing a plurality of pieces of biological information to which identification information is assigned and a plurality of pieces of biological information to which no identification information is given,
The biological information input unit includes a biological detection unit that detects a biological body,
When the biometric information input unit acquires biometric information without the identification information input unit acquiring the identification information, the biometric authentication verification is performed using the acquired biometric information and the biometric information stored in the verification database. Department collates,
When the identification information input unit acquires identification information and the biometric information input unit acquires biometric information, the biometric information stored in the collation database corresponding to the acquired identification information and the acquired and the biological information, the control is such that the biometric authentication checking unit is checking to determine authentication success or failure,
When the living body detection unit detects a living body,
The control unit causes the biometric information input unit to operate a process of reading biometric information,
A biometric authentication system that operates to read a barcode when a biometric is not detected. When the identification information input unit does not acquire identification information and the biological information input unit acquires biological information,
Obtaining unmatched biometric information feature quantities from the matching database, collating with the biometric information feature quantities of the obtained biometric information,
When the identification information input unit acquires identification information and the biological information input unit acquires biological information,
The biometric authentication system according to claim 1, wherein a biometric feature quantity corresponding to the obtained identification information is obtained from the collation database and collated with the biometric feature quantity of the obtained biometric information. A biometric information input unit for inputting biometric information of the user;
An identification information input unit for inputting user identification information;
A biometric authentication verification unit that performs verification processing using the biometric information of the user;
A control unit that controls the biometric verification unit based on an input from the biometric information input unit or the identification information input unit;
A collation database storing a plurality of pieces of biological information to which identification information is assigned and a plurality of pieces of biological information to which no identification information is given,
The collation database stores a first collation database in which a plurality of pieces of biometric information are stored, and a plurality of pieces of biometric information having lower quality than the biometric information stored in the first collation database in association with identification information. Divided into a second verification database,
The biometric verification unit
When the identification information input unit does not acquire identification information and the biological information input unit acquires biological information,
Collating the acquired biometric information with the biometric information stored in the first collation database;
When the identification information input unit acquires identification information and the biological information input unit acquires biological information,
A biometric authentication system, wherein biometric information stored in the second collation database corresponding to the acquired identification information is collated with the acquired biometric information. When the identification information input unit does not acquire identification information and the biological information input unit acquires biological information,
Obtaining an unmatched biometric information feature from the first matching database, collating it with the biometric feature of the obtained biometric information,
When the identification information input unit acquires identification information and the biological information input unit acquires biological information,
The biometric authentication system according to claim 3, wherein a biometric information feature amount corresponding to the acquired identification information is acquired from the second collation database and collated with the biometric feature amount of the acquired biometric information. . The control unit includes an input information determination unit that determines whether the identification information and the biological information are input;
When the input determination unit acquires identification information, the biometric verification collating unit collates the acquired biometric information with the biometric information stored in the collation database,
When the input information determination unit has not acquired identification information, the biometric authentication verification unit displays the biometric information stored in the verification database corresponding to the acquired identification information and the acquired biometric information. The biometric authentication system according to claim 3, wherein the biometric authentication system is controlled so as to be collated. The biological information input unit includes a biological detection unit that detects a biological body,
When the living body detection unit detects a living body,
The control unit causes the biometric information input unit to operate a process of reading biometric information,
The biometric authentication system according to any one of claims 3 to 5, wherein when a biometric is not detected, a process of reading a barcode is operated. In the case where the biometric information input unit acquires the biometric information of the user without the identification information input unit acquiring the user's identification information, the acquired biometric information and the first collation database are stored. If collation fails as a result of collation with biometric information, the user is prompted to enter identification information,
The biometric information is extracted from the first collation database or the second collation database based on the input identification information, and collation processing is performed. Biometric authentication system. In the case where the biometric information input unit acquires the biometric information of the user without the identification information input unit acquiring the user's identification information, the acquired biometric information and the first collation database are stored. If matching fails with biometric information,
Using the acquired biometric information, rematching with the biometric information stored in the second collation database,
If the verification is successful, prompt the user for identification information,
The biometric authentication system according to any one of claims 3 to 6, wherein collation is performed between the input identification information and identification information corresponding to the biometric information that has been successfully collated. The biometric authentication system has a biometric information registration terminal,
The biometric information registration terminal
A biometric information acquisition unit that acquires biometric information from the user;
A control unit that extracts biometric information features from the acquired biometric information;
A quality determination unit that determines the quality of the extracted biometric feature amount,
The controller is
If the quality of the biometric information feature quantity is equal to or higher than a specified value, the biometric information feature quantity is stored in the first verification database;
9. The biometric information feature quantity is stored in the second collation database in association with user identification information when the quality of the biometric feature quantity is less than a prescribed value. The biometric authentication system as described in any one.   The biometric authentication system according to claim 9, wherein the quality determination unit determines the quality based on the number of pixels identified as a blood vessel in the acquired biometric information. The control unit includes an input information determination unit that determines whether the identification information and the biological information are input;
When the input determination unit acquires identification information, the biometric verification collating unit collates the acquired biometric information with the biometric information stored in the collation database,
When the input information determination unit has not acquired identification information, the biometric authentication verification unit displays the biometric information stored in the verification database corresponding to the acquired identification information and the acquired biometric information. The biometric authentication system according to claim 1, wherein the biometric authentication system is controlled so as to be collated. The collation database is a first collation database in which a plurality of pieces of biometric information is stored, and biometric information having a quality lower than that of the biometric information stored in the first collation database is stored in association with identification information. Divided into two collation databases,
When the identification information input unit acquires the biological information without acquiring the identification information,
Obtaining an unmatched biometric information feature from the first matching database, collating it with the biometric feature of the obtained biometric information,
When the biometric information input unit acquires biometric information and the identification information input unit acquires identification information,
The biometric information feature amount corresponding to the acquired identification information is acquired from the second collation database, and the biometric information feature amount of the acquired biometric information is collated to determine whether authentication is successful or unsuccessful. Biometric authentication method.

Images