# US 2019/0377934 A1 · Face recognition system and method for enhancing face recognition

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

- Number: US 2019/0377934 A1
- Document ID: US20190377934A1
- Status: Published application
- Kind: A1
- Application number: 16/431,513
- Filed: 2019-06-04
- Published: 2019-12-12
- Earliest claimed: 2018-06-08
- Assignees: Pegatron Corporation
- Applicants: Pegatron Corporation
- Inventors: YU-HUNG TSENG
- Web page: https://patentgrep.com/patent/US20190377934A1

## CPC

- G06V 40/168: {Feature extraction; Face representation}
- G06V 10/772: Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
- G06V 40/16: Human faces, e.g. facial parts, sketches or expressions
- G06V 40/172: {Classification, e.g. identification}
- H04L 67/10: in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A face recognition system and a method for enhancing face recognition are provided. The method includes: receiving a face image and obtaining a feature of the face image from a feature extraction model; registering the face image to set the feature of the face image as a first recognition feature; performing a synthesis operation on the face image according to at least one first adjustment parameter to generate a synthetic image, and obtaining a feature of the synthetic image from the feature extraction model; comparing first recognition feature with the feature of the synthetic image to obtain a feature similarity; comparing the feature similarity with a threshold value to obtain a comparison result; and registering the synthetic image when the comparison result indicates that the feature similarity is less than or equal to the threshold value.

## Claims

### Claim 1

A method for enhancing face recognition, comprising:
receiving a face image, and obtaining a feature of the face image from a feature extraction model;
registering the face image to set the feature of the face image as a first recognition feature;
performing a synthesis operation on the face image according to at least one first adjustment parameter to generate a synthetic image, and obtaining a feature of the synthetic image from the feature extraction model;
comparing the first recognition feature with the feature of the synthetic image to obtain a feature similarity;
comparing the feature similarity with a threshold value to obtain a comparison result, wherein the threshold value is generated according to a face recognition error rate; and
registering the synthetic image when the comparison result indicates that the feature similarity is less than or equal to the threshold value.

### Claim 2 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein registering the face image comprises:
storing the first recognition feature and the face image.

### Claim 3 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein the at least one first adjustment parameter is generated by use of at least one of a Monte Carlo algorithm and a genetic algorithm.

### Claim 4 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein the synthesis operation is to change a contrast, brightness and simulate rotating angles of the face image according to the at least one first adjustment parameter.

### Claim 5 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein the synthesis operation is to cover a part of area of the face image with a cover image according to the at least one first adjustment parameter.

### Claim 6 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein the step of registering the synthetic image comprises:
setting the feature of the synthetic image as a second recognition feature; and
storing the second recognition feature.

### Claim 7 (depends on 1)

The method for enhancing face recognition according to claim 1, further comprising:
when the comparison result indicates that the feature similarity is higher than the threshold value, performing the synthesis operation on the face image according to at least one second adjustment parameter to generate another synthetic image.

### Claim 8 (depends on 1)

The method for enhancing face recognition according to claim 1, wherein the face recognition error rate approximates zero.

### Claim 9

A face recognition system, comprising:
an image extraction device, configured to extract a face image;
a feature extraction model, coupled to the image extraction device and configured to receive the face image and obtain a feature of the face image; and
an image processing unit, coupled to the image extraction device and the feature extraction model, wherein the image processing unit is configured to:
register the face image to set the feature of the face image as a first recognition feature;
perform a synthesis operation on the face image according to at least one first adjustment parameter to generate a synthetic image and obtain a feature of the synthetic image from the feature extraction model;
compare the first recognition feature with the feature of the synthetic image to obtain a feature similarity;
compare the feature similarity with a threshold value to obtain a comparison result, wherein the threshold value is generated according to a face recognition error rate; and
register the synthetic image when the comparison result indicates that the feature similarity is less than or equal to the threshold value.

### Claim 10 (depends on 9)

The face recognition system according to claim 9, further comprising:
a storage device, coupled to the image processing unit, the storage device being configured to store the first recognition feature and the face image.

### Claim 11 (depends on 9)

The face recognition system according to claim 9, wherein the at least one first adjustment parameter is generated by use of at least one of a Monte Carlo algorithm and a genetic algorithm.

### Claim 12 (depends on 9)

The face recognition system according to claim 9, wherein the synthesis operation is to change a contrast, brightness and simulate rotating angles of the face image according to the at least one first adjustment parameter.

### Claim 13 (depends on 9)

The face recognition system according to claim 9, wherein the synthesis operation is to cover a part of area of the face image with a cover image according to the at least one first adjustment parameter.

### Claim 14 (depends on 10)

The face recognition system according to claim 10, wherein the operation that the image processing unit registers the synthetic image comprises that the image processing unit sets the feature of the synthetic image as a second recognition feature and stores the second recognition feature in the storage device.

### Claim 15 (depends on 9)

The face recognition system according to claim 9, wherein, when the comparison result indicates that the feature similarity is higher than the threshold value, the image processing unit performs a synthesis operation on the face image according to at least one second adjustment parameter to generate another synthetic image.

### Claim 16 (depends on 9)

The face recognition system according to claim 9, wherein the face recognition error rate approximates zero.

### Claim 17 (depends on 9)

The face recognition system according to claim 9, wherein the feature extraction model is integrated in the image processing unit or integrated in a cloud server.
