Identification apparatus, identification method, and program

• US 9,129,145 B2
• Filed: 09/11/2012
• Issued: 09/08/2015
• Est. Priority Date: 03/19/2010
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Abiko, Yukihiro
• Primary Examiner(s): AHMED, SAMIR ANWAR
• Application Number:
  US13/610,224
• Publication Number:
  US 20130004032A1
• Time in Patent Office: 1,092 Days
• Field of Search: 382/115, 382/124, 356/71, 283 68- 70
• US Class Current: 1/1
• CPC Class Codes: G06V 40/1365 Matching; Classification

Abstract

An identification apparatus includes a classification unit that determines two or more classes into which input biometric data is classified out of a plurality of classes based on features of the input biometric data, where a plurality of items of registered biometric data have been classified into at least one of the plurality of classes, a calculation unit that calculates similarity between the input biometric data and each item of the registered biometric data registered in each of the two or more classes into which the input biometric data is classified, and an identification unit that identifies data on a user who has entered the input biometric data among the registered biometric data registered in any of the two or more classes into which the input biometric data is classified, based on the similarity to the input biometric data.

Claims

1. An identification apparatus comprising a processor executing a process that causes the identification apparatus to perform:

determining two or more classes into which input biometric data is classified out of a plurality of classes based on features of the input biometric data, where a plurality of items of registered biometric data have been classified into at least one of the plurality of classes;

calculating similarity between the input biometric data and each item of the registered biometric data registered in each of the determined two or more classes into which the input biometric data is classified; and

identifying data on a user who has entered the input biometric data among the registered biometric data registered in any of the determined two or more classes, based on the similarity to the input biometric data; and

setting a threshold for each of the determined two or more classes in order to identify data on the user who has entered the input biometric data, wherein

each of the classes is set in advance,

when, among the registered biometric data registered in each of the determined two or more classes, there is at least one biometric data item whose similarity to the input biometric data exceeds the threshold set for each of the determined two or more classes, the identifying identifies that the input biometric data and the registered biometric data item belong to the same user.

2. The identification apparatus according to claim 1, wherein the setting sets the threshold for each of the determined two or more classes based on a false accept rate expected of the identification apparatus and a false accept rate expected of each of the determined two or more classes.

3. The identification apparatus according to claim 1, wherein the processor executes the process that further causes the identification apparatus to perform registering the biometric data entered during registration into the determined two or more classes determined by the determining.

4. The identification apparatus according to claim 1, wherein the processor executes the process that further causes the identification apparatus to perform extracting at least one partial region from the input biometric data,

wherein, based on a feature included in the at least one partial region, the determining determines two or more classes into which the at least one partial region is classified, and

the calculating calculates similarity between the at least one partial region and the partial region registered in each of the determined two or more classes into which the at least one partial region is classified.

5. The identification apparatus according to claim 1, wherein the processor executes the process that further causes the identification apparatus to perform:

detecting an input state of the input biometric data;

estimating a method for improving the input state of the input biometric data; and

notifying the user of the method for improving the input state of the input biometric data.

6. An identification method executed by a computer, comprising:

determining two or more classes into which input biometric data is classified out of a plurality of classes based on features of the input biometric data, where a plurality of items of registered biometric data have been classified into at least one of the plurality of classes;

calculating similarity between the input biometric data and each item of the registered biometric data registered in each of the determined two or more classes into which the input biometric data is classified; and

identifying data on a user who has entered the input biometric data among the registered biometric data registered in any of the determined two or more classes, based on the similarity to the input biometric data; and

setting a threshold for each of the determined two or more classes in order to identify data on the user who has entered the input biometric data, wherein

each of the classes is set in advance,

when, among the registered biometric data registered in each of the determined two or more classes, there is at least one biometric data item whose similarity to the input biometric data exceeds the threshold set for each of the determined two or more classes, the identifying identifies that the input biometric data and the registered biometric data item belong to the same user.

7. A non-transitory computer-readable recording medium recording thereon a program executed by a computer, the program comprising:

determining two or more classes into which input biometric data is classified out of a plurality of classes based on features of the input biometric data, where a plurality of items of registered biometric data have been classified into at least one of the plurality of classes;

calculating similarity between the input biometric data and each item of the registered biometric data registered in each of the determined two or more classes into which the input biometric data is classified; and

identifying data on a user who has entered the input biometric data among the registered biometric data registered in any of the determined two or more classes, based on the similarity to the input biometric data; and

setting a threshold for each of the determined two or more classes in order to identify data on the user who has entered the input biometric data, wherein

each of the classes is set in advance,

when, among the registered biometric data registered in each of the determined two or more classes, there is at least one biometric data item whose similarity to the input biometric data exceeds the threshold set for each of the determined two or more classes, the identifying identifies that the input biometric data and the registered biometric data item belong to the same user.