# US 2021/0174067 A1 · Live facial recognition system and method

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

- Number: US 2021/0174067 A1
- Document ID: US20210174067A1
- Status: Published application
- Kind: A1
- Application number: 16/777,514
- Filed: 2020-01-30
- Published: 2021-06-10
- Earliest claimed: 2019-12-10
- Assignees: Wistron Corporation
- Applicants: Wistron Corporation
- Inventors: Yao-Tsung Chang; Chuan-Yen Kao; Chih-Yang Hung
- Web page: https://patentgrep.com/patent/US20210174067A1

## CPC

- G06V 40/172: {Classification, e.g. identification}
- G06T 7/564: from contours
- G06T 7/90: Determination of colour characteristics
- G06V 40/162: {using pixel segmentation or colour matching}
- G06V 40/168: {Feature extraction; Face representation}
- G06V 40/45: {Detection of the body part being alive}
- G06T 2207/10004: Still image; Photographic image
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A live facial recognition method includes capturing a zoom-out image of a face of a subject under recognition; and detecting a frame outside the face of the subject under recognition on the zoom-out image. The subject under recognition is determined to be a living subject when the zoom-out image includes no frame outside the face.

## Claims

### Claim 1

A live facial recognition method, comprising:
(a) capturing a zoom-out image of a face of a subject under recognition; and
(b) detecting a frame outside the face of the subject under recognition on the zoom-out image;
wherein the subject under recognition is determined to be a living subject when the zoom-out image includes no frame outside the face.

### Claim 2 (depends on 1)

The method of claim 1, wherein the zoom-out image is captured after adjusting a field of view of an image capture device.

### Claim 3 (depends on 1)

The method of claim 1, wherein the step (b) comprises:
transforming pixels of the zoom-out image to obtain a transformed image;
comparing pixels of the transformed image with a predetermined threshold to obtain a contour image; and
detecting the frame outside the face of the subject under recognition according to the contour image.

### Claim 4 (depends on 3)

The method of claim 3, wherein the threshold is obtained by analyzing energy distribution of the zoom-out image.

### Claim 5 (depends on 1)

The method of claim 1, wherein the step (b) comprises:
calculating a sum of squares or root mean square of color difference between neighboring pixels of the zoom-out image to obtain a contour image; and
detecting the frame outside the face of the subject under recognition according to the contour image.

### Claim 6 (depends on 1)

The method of claim 1, wherein the frame comprises a closed quadrilateral.

### Claim 7 (depends on 1)

The method of claim 1, wherein the frame comprises at least two edges of a quadrilateral, and the at least two edges and boundaries of the zoom-out image construct a closed quadrilateral.

### Claim 8 (depends on 1)

The method of claim 1, before the step (a), further comprising:
capturing at least one image of the face of the subject under recognition;
extracting at least one feature data according to the at least one image; and
comparing the feature data with a facial feature database.

### Claim 9

A live facial recognition system, comprising:
an image capture device that captures a zoom-out image of a face of a subject under recognition;
a frame recognition module that detects a frame outside the face of the subject under recognition on the zoom-out image; and
an output module that determines the subject under recognition to be a living subject when the zoom-out image includes no frame outside the face.

### Claim 10 (depends on 9)

The system of claim 9, wherein the image capture device comprises a camera, wherein the zoom-out image is captured after adjusting a field of view of the camera.

### Claim 11 (depends on 9)

The system of claim 9, further comprising:
a facial recognition module that receives at least one image of the face of the subject under recognition, extracts at least one feature data according to the at least one image, and compares the feature data with a facial feature database.

### Claim 12 (depends on 11)

The system of claim 11, wherein the image capture device comprises a first camera and a second camera, the second camera having a field of view larger than the first camera, wherein the first camera captures the at least one image for the facial recognition module and the second camera captures the zoom-out image for the frame recognition module.

### Claim 13 (depends on 9)

The system of claim 9, wherein the frame recognition module comprises:
an image transformation device that transforms pixels of the zoom-out image to obtain a transformed image;
a binary thresholding device that compares pixels of the transformed image with a predetermined threshold to obtain a contour image; and
a frame detection device that detects the frame outside the face of the subject under recognition according to the contour image.

### Claim 14 (depends on 13)

The system of claim 13, wherein the frame recognition module further comprises:
an energy analysis device that analyzes energy distribution of the zoom-out image to obtain the predetermined threshold.

### Claim 15 (depends on 9)

The system of claim 9, wherein the frame recognition module comprises:
an image transformation device that calculates a sum of squares or root mean square of color difference between neighboring pixels of the zoom-out image to obtain a contour image; and
a frame detection device that detects the frame outside the face of the subject under recognition according to the contour image.

### Claim 16 (depends on 9)

The system of claim 9, wherein the frame comprises a closed quadrilateral.

### Claim 17 (depends on 9)

The system of claim 9, wherein the frame comprises at least two edges of a quadrilateral, and the at least two edges and boundaries of the zoom-out image construct a closed quadrilateral.
