HYBRID BIOMETRIC AUTHENTICATION DEVICE, HYBRID BIOMETRIC AUTHENTICATION METHOD, AND COMPUTER-READABLE MEDIUM STORING HYBRID BIOMETRIC AUTHENTICATION PROGRAM

• US 20100284575A1
• Filed: 07/19/2010
• Published: 11/11/2010
• Est. Priority Date: 01/29/2008
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha DDS
• Original Assignee: Kabushiki Kaisha DDS
• Inventors: Umezaki, Taizo, Yoshimine, Tatsuki, Hoguro, Masahiro, Matsuo, Masaaki
• Granted Patent: US 8,369,583 B2
• Time in Patent Office: Days
• US Class Current: 382/116
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

A hybrid biometric authentication device that includes a plurality of feature extraction portions that respectively extract, from biometric information, a plurality of types of feature information that have mutually distinct properties, a feature combining portion that generates hybrid feature information by combining the plurality of types of feature information, a feature information storage portion that stores the hybrid feature information, a plurality of similarity computation portions that separates each of new information and registered information into the plurality of types of feature information, and compute a plurality of degrees of similarity between separate sets of a same type of the feature information, and a determination portion that, based on the plurality of degrees of similarity, determines whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from a same person.

Claims

1. A hybrid biometric authentication device, comprising:

a plurality of feature extraction portions that respectively extract, from biometric information, a plurality of types of feature information that have mutually distinct properties;

a feature combining portion that generates hybrid feature information by combining the plurality of types of feature information respectively extracted by the plurality of the feature extraction portions;

a feature information storage portion that stores the hybrid feature information generated by the feature combining portion;

a plurality of similarity computation portions that separate each of new information and registered information into the plurality of types of feature information before combined, and compute a plurality of degrees of similarity between separate sets of a same type of the feature information that have been respectively separated from the new information and the registered information, the new information being the hybrid feature information newly generated by the feature combining portion, the registered information being the hybrid feature information stored in advance in the feature information storage portion; and

a determination portion that, based on the plurality of degrees of similarity computed by the plurality of similarity computation portions, determines whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from a same person.

2. The hybrid biometric authentication device according to claim 1, wherein

the plurality of feature extraction portions and the plurality of similarity computation portions each use a corresponding plurality of algorithms that have mutually distinct properties and different authentication performance.

3. The hybrid biometric authentication device according to claim 1, wherein

the plurality of similarity computation portions, in a case where at least one of the plurality of types of feature information extracted by the plurality of feature extraction portions from the biometric information input from a biometric information input device is invalid for at least corresponding one of the plurality of algorithms, compute the plurality of degrees of similarity using information that indicates that the at least one of the plurality types of feature information is invalid.

4. The hybrid biometric authentication device according to claim 1, wherein

the feature combining portion combines the plurality of types of feature information after removing information included in all of the plurality of types of feature information respectively extracted by the plurality of feature extraction portions, then generates one set of the hybrid feature information by compressing and encrypting the combined information in accordance with a system application.

5. The hybrid biometric authentication device according to claim 4, wherein

the plurality of similarity computation portions respectively decrypt and decompress the new information and the registered information, then respectively compute the plurality of degrees of similarity between the separate sets of the same type of feature information among the plurality of types of feature information.

6. The hybrid biometric authentication device according to claim 1, wherein

the determination portion includes:

an overall degree of similarity computation portion that, in accordance with a standard degree of similarity that is one of the plurality of degrees of similarity computed by the plurality of the degree of similarity computation portions and in accordance with another of the plurality of the degrees of similarity that is not the standard degree of similarity, computes a single overall degree of similarity by one of increasing and decreasing a value of the standard degree of similarity; and

an overall determination portion that, based on the overall degree of similarity computed by the overall similarity computation portion, determines whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from the same person.

7. The hybrid biometric authentication device according to claim 6, wherein

the determination portion determines that the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from the same person in a case where the overall degree of similarity computed by the overall degree of similarity computation portion is not less than a specified threshold value, and determines that the biometric information on which the new information is based and the biometric information on which the registered information is based have not been acquired from the same person in a case where the overall degree of similarity is less than the specified threshold value.

8. A hybrid biometric authentication method, comprising the steps of:

extracting, from biometric information, a plurality of types of feature information that have mutually distinct properties;

generating hybrid feature information by combining the plurality of the types of feature information that have been extracted;

separating each of new information and registered information into the plurality of types of feature information before combined, and computing a plurality of degrees of similarity between separate sets of a same type of the feature information that have been respectively separated from the new information and the registered information, the new information being the hybrid feature information that has been newly generated, the registered information being the hybrid feature information that has been generated and stored in advance in a storage portion; and

determining whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from a same person, based on the plurality of the degrees of similarity that have been computed.

9. The hybrid biometric authentication method according to claim 8, wherein

a plurality of algorithms that have mutually distinct properties and different authentication performance are used both in the step of extracting the plurality of types of feature information and in the step of computing the plurality of degrees of similarity.

10. The hybrid biometric authentication method according to claim 8, wherein

in the step of computing the plurality of degrees of similarity, the plurality of degrees of similarity are computed, in a case where at least one of the plurality of types of feature information that have been extracted from the biometric information that has been input from a biometric information input device is invalid for at least one of the plurality of algorithms, using information that indicates that the at least one of the plurality of types of feature information is invalid.

11. The hybrid biometric authentication method according to claim 8, wherein

in the step of generating the hybrid feature information, one set of the hybrid feature information is generated by combining the plurality of types of feature information after removing information that is included in all of the plurality of types of feature information that have been extracted, then compressing and encrypting the combined information in accordance with a system application.

12. The hybrid biometric authentication method according to claim 11, wherein

in the step of computing the plurality of degrees of similarity, the new information and the registered information are decrypted and decompressed, and the plurality of the degrees of similarity between the separate sets of the same type of feature information among the plurality of types of feature information are then computed.

13. The hybrid biometric authentication method according to claim 8, wherein

in the step of determining based on the plurality of degrees of similarity, a single overall degree of similarity is computed by one of increasing and decreasing a value of a standard degree of similarity that is one of the plurality of degrees of similarity that have been computed, in accordance with the standard degree of similarity and in accordance with another of the plurality of degrees of similarity that is not the standard degree of similarity, and then the determination is made, based on the overall degree of similarity, as to whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from the same person.

14. The hybrid biometric authentication method according to claim 13, wherein

in the step of determining based on the plurality of degrees of similarity, a determination is made that the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from the same person in a case where the overall degree of similarity is not less than a specified threshold value, and a determination is made that the biometric information on which the new information is based and the biometric information on which the registered information is based have not been acquired from the same person in a case where the overall degree of similarity is less than the specified threshold value.

15. A computer-readable medium storing a hybrid biometric authentication program, the program comprising instructions that cause a computer to perform the steps of:

extracting, from biometric information, a plurality of types of feature information that have mutually distinct properties;

generating hybrid feature information by combining the plurality of the types of feature information that have been extracted;

separating each of new information and registered information into the plurality of types of feature information before combined, and computing a plurality of degrees of similarity between separate sets of a same type of the feature information that have been respectively separated from the new information and the registered information, the new information being the hybrid feature information that has been newly generated, the registered information being the hybrid feature information that has been generated and stored in advance in a storage portion; and

determining whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from a same person, based on the plurality of the degrees of similarity that have been computed.

16. The computer-readable medium according to claim 15, wherein

a plurality of algorithms that have mutually distinct properties and different authentication performance are used both in the step of extracting the plurality of types of feature information and in the step of computing the plurality of degrees of similarity.

17. The computer-readable medium according to claim 16, wherein

in the step of computing the plurality of degrees of similarity, the plurality of degrees of similarity are computed, in a case where at least one of the plurality of types of feature information that have been extracted from the biometric information that has been input from a biometric information input device is invalid for at least one of the plurality of algorithms, using information that indicates that the at least one of the plurality of types of feature information is invalid.

18. The computer-readable medium according to claim 15, wherein

in the step of generating the hybrid feature information, one set of the hybrid feature information is generated by combining the plurality of types of feature information after removing information that is included in all of the plurality of types of feature information that have been extracted, then compressing and encrypting the combined information in accordance with a system application.

19. The computer-readable medium according to claim 18, wherein

in the step of computing the plurality of degrees of similarity, the new information and the registered information are decrypted and decompressed, and the plurality of the degrees of similarity between the separate sets of the same type of feature information among the plurality of types of feature information are then computed.

20. The computer-readable medium according to claim 15, wherein

in the step of determining based on the plurality of degrees of similarity, a single overall degree of similarity is computed by one of increasing and decreasing a value of a standard degree of similarity that is one of the plurality of degrees of similarity that have been computed, in accordance with the standard degree of similarity and in accordance with another of the plurality of degrees of similarity that is not the standard degree of similarity, and then the determination is made, based on the overall degree of similarity, as to whether the biometric information on which the new information is based and the biometric information on which the registered information is based have been acquired from the same person.