# US 2022/0036040 A1 · Biometric recognition apparatus and biometric recognition method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0036040 A1
- Document ID: US20220036040A1
- Status: Published application
- Kind: A1
- Application number: 17/502,925
- Filed: 2021-10-15
- Published: 2022-02-03
- Earliest claimed: 2017-10-13
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Tatsuya Shimahara
- Web page: https://patentgrep.com/patent/US20220036040A1

## CPC

- G06V 40/1353: {Extracting features related to minutiae or pores}
- G06F 18/2115: by evaluating different subsets according to an optimisation criterion, e.g. class separability, forward selection or backward elimination
- G06V 10/771: Feature selection, e.g. selecting representative features from a multi-dimensional feature space
- G06V 40/1347: {Preprocessing; Feature extraction}
- G06V 40/1365: {Matching; Classification}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A biometric recognition apparatus and fingerprint feature extraction method can automatically optimize parameters used for extracting a feature template from a biometric image. The biometric recognition apparatus includes: a teacher data generation unit that generates a genuine pair and an imposter pair of first and second biometric images; a learning data generation unit that uses a plurality of different temporary parameters to extract feature templates from the first biometric image and the second biometric image; and an optimum solution determination unit that calculates a score separation degree on the temporary parameter basis based on a first score representing a similarity degree of a pair of the feature templates extracted from the genuine pair and a second score representing a similarity degree of a pair of the feature templates extracted from the imposter pair and determines the temporary parameter based on a level of the score separation degree.

## Claims

### Claim 1

A biometric recognition apparatus comprising:
a teacher data generation unit that generates, from a plurality of first biometric images and a plurality of second biometric images, a plurality of pairs each of which is a combination of the first biometric image and the second biometric image;
a learning data generation unit that uses a plurality of different temporary parameters to extract feature templates from the first biometric image and the second biometric image; and
a calculation unit that calculates a similarity degree of the feature templates extracted from the first biometric image and the second biometric image that become the pair, for the plurality of pairs generated by the teacher data generation unit; and
an optimum solution determination unit that determines the temporary parameter based on a calculation result by the calculation unit.

### Claim 2 (depends on 1)

The biometric recognition apparatus according to claim 1,
wherein the teacher data generation unit generates a genuine pair and an imposter pair from the plurality of first biometric images and the plurality of second biometric images,
wherein the calculation unit calculates a score separation degree on the temporary parameter basis based on a first score representing a similarity degree of the feature templates extracted from the first biometric image and the second biometric image that become the genuine pair and a second score representing a similarity degree of the feature templates extracted from the first biometric image and the second biometric image that become the imposter pair, and
wherein the optimum solution determination unit determines the temporary parameter based on a level of the score separation degree.

### Claim 3 (depends on 2)

The biometric recognition apparatus according to claim 2, wherein the optimum solution determination unit determines the temporary parameter based on the score separation degree which is the largest.

### Claim 4 (depends on 2)

The biometric recognition apparatus according to claim 2, wherein the calculation unit calculates the first score and the second score for each of the first biometric image or the second biometric image, and calculates, as the score separation degree, a ratio of the first score ranked top.

### Claim 5 (depends on 2)

The biometric recognition apparatus according to claim 2, wherein the calculation unit removes the genuine pair having the first score greater than or equal to a predetermined value from a calculation process of the score separation degree.

### Claim 6 (depends on 1)

The biometric recognition apparatus according to claim 1 further comprising:
a storage unit that stores a parameter used for extracting the feature templates from the first biometric image and the second biometric image; and
a parameter update unit that updates the parameter stored in the storage unit with the determined temporary parameter.

### Claim 7 (depends on 1)

The biometric recognition apparatus according to claim 1,
wherein a pair of the first biometric image and the second biometric image is a pair of collected images of a fingerprint or a palm print, and
wherein the feature templates form a feature point group of a fingerprint or a palm print.

### Claim 8 (depends on 7)

The biometric recognition apparatus according to claim 7, wherein a pair of the first biometric image and the second biometric image is a pair of a collected image of a latent fingerprint and a collected image of a tenprint fingerprint.

### Claim 9 (depends on 1)

The biometric recognition apparatus according to claim 1, wherein the temporary parameter includes a combination of a noise reduction process and an enhancement process and a weight.

### Claim 10

A biometric recognition method used in a control calculation unit of a biometric recognition apparatus, the biometric recognition method comprising:
a teacher data generation step of generating, from a plurality of first biometric images and a plurality of second biometric images, a plurality of pairs each of which is a combination of the first biometric image and the second biometric image;
a learning data generation step of using a plurality of different temporary parameters to extract feature templates from the first biometric image and the second biometric image; and
a calculation step of calculating a similarity degree of the feature templates extracted from the first biometric image and the second biometric image that become the pair, for the plurality of pairs generated by the teacher data generation step; and
an optimum solution determination step of determining the temporary parameter based on a calculation result by the calculation step.

### Claim 11

A computer readable non-transitory storage medium storing a program that causes a computer to function as:
a teacher data generation unit configured to generate, from a plurality of first biometric images and a plurality of second biometric images, a plurality of pairs of which is a combination of the first biometric image and the second biometric image;
a learning data generation unit configured to use a plurality of different temporary parameters to extract feature templates from the first biometric image and the second biometric image; and
a calculation unit configured to calculate a similarity degree of the feature templates extracted from the first biometric image and the second biometric image that become the pair, for the plurality of pairs generated by the teacher data generation unit; and
an optimum solution determination unit configured to determine the temporary parameter based on a calculation result by the calculation unit.
