FINGERPRINT AUTHENTICATION SYSTEM, FINGERPRINT AUTHENTICATION METHOD, AND FINGERPRINT AUTHENTICATION PROGRAM

• US 20130051638A1
• Filed: 05/18/2011
• Published: 02/28/2013
• Est. Priority Date: 06/04/2010
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Monden, Akira
• Granted Patent: US 8,942,439 B2
• Time in Patent Office: Days
• US Class Current: 382/125
• CPC Class Codes: A61B 5/1172 using fingerprinting
• CPC Class Codes: G06V 40/1359 Extracting features related to ridge properties; Determining the fingerprint type, e.g. whorl or loop

Abstract

To improve the precision with which a false finger formed by attaching a transparent thin film to a surface of a finger is identified, a fingerprint authentication system includes: an image division unit 31 that divides a captured image of an authentication subject to be subjected to fingerprint authentication into a plurality of small regions; a frequency analysis unit 32 that performs a frequency analysis on each of the small regions; a ridge candidate extraction unit 33 that extracts frequency components corresponding to ridge candidates in the small regions from frequency components obtained in the frequency analysis; a continuous ridge candidate group generation unit 34 that compares the ridge candidates between adjacent small regions, and when the ridge candidates satisfy a condition enabling the ridge candidates to be determined continuous, generates a continuous ridge candidate group including the ridge candidates; and a determination unit 35 that determines that an abnormality exists in the image when overlap exists between the image regions corresponding to the continuous ridge candidate group.

Claims

1. A fingerprint authentication system comprising:

an image division unit that divides a captured image of an authentication subject to be subjected to fingerprint authentication into a plurality of small regions;

a frequency analysis unit that performs a frequency analysis on each of the small regions;

a ridge candidate extraction unit that extracts frequency components corresponding to ridge candidates in the small regions from frequency components obtained in the frequency analysis;

a continuous ridge candidate group generation unit that compares the ridge candidates between adjacent small regions, and when the ridge candidates satisfy a condition enabling the ridge candidates to be determined continuous, generates a continuous ridge candidate group including the ridge candidates; and

a determination unit that determines that an abnormality exists in the image when overlap exists between the image regions corresponding to the continuous ridge candidate group.

2. The fingerprint authentication system according to claim 1, wherein the frequency analysis unit performs a two-dimensional Fourier transform as the frequency analysis, and

the ridge candidate extraction unit extracts a group of local maximum points having origin symmetry when expressed on a frequency space by the two-dimensional Fourier transform as the frequency components corresponding to the ridge candidates.

3. The fingerprint authentication system according to claim 1, wherein the condition enabling the ridge candidates to be determined continuous is satisfied when a difference or a ratio between values of at least a part of a direction, a width, an amplitude, and a phase of the ridge candidates is smaller than a predetermined value.

4. The fingerprint authentication system according to claim 1, wherein the continuous ridge candidate group generation unit ensures that a single ridge candidate is not included in a plurality of continuous ridge candidate groups.

5. The fingerprint authentication system according to claim 1, wherein the determination unit determines that the abnormality exists in the image when a size of the overlap between the image regions corresponding to the continuous ridge candidate group equals or exceeds a predetermined size.

6. The fingerprint authentication system according to claim 1, wherein the abnormality corresponds to a case in which an image of a false finger formed by attaching a thin film to a surface of a finger is captured.

7. The fingerprint authentication system according to claim 1, wherein the ridge candidate extraction unit includes a first ridge candidate extraction unit that extracts ridge candidates determined to have a large amplitude, and a second ridge candidate extraction unit that extracts ridge candidates determined to have a small amplitude,

the continuous ridge candidate group generation unit includes a first continuous ridge candidate group generation unit that generates a first continuous ridge candidate group by comparing the ridge candidates extracted by the first ridge candidate extraction unit, and a second continuous ridge candidate group generation unit that generates a second continuous ridge candidate group by comparing the ridge candidates extracted by the second ridge candidate extraction unit, and

the determination unit determines that the abnormality exists in the image when overlap exists between an image region corresponding to the first continuous ridge candidate group and an image region corresponding to the second continuous ridge candidate group, or when overlap exists between the image regions corresponding to the second continuous ridge candidate group.

8. A fingerprint authentication method comprising the steps of:

dividing a captured image of an authentication subject to be subjected to fingerprint authentication into a plurality of small regions;

performing a frequency analysis on each of the small regions;

extracting frequency components corresponding to ridge candidates in the small regions from frequency components obtained in the frequency analysis;

comparing the ridge candidates between adjacent small regions, and when the ridge candidates satisfy a condition enabling the ridge candidates to be determined continuous, generating a continuous ridge candidate group including the ridge candidates; and

determining that an abnormality exists in the image when overlap exists between the image regions corresponding to the continuous ridge candidate group.

9. A fingerprint authentication program for causing a computer to execute the steps according to claim 8.