# US 10,441,198 B2 · Face detection device, face detection system, and face detection method

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

- Number: US 10,441,198 B2
- Document ID: US10441198B2
- Status: Granted
- Kind: B2
- Application number: 15/580,809
- Filed: 2016-05-10
- Published: 2018-08-09
- Granted: 2019-10-15
- Earliest claimed: 2015-06-19
- Assignees: Panasonic Intellectual Property Management Co., Ltd.
- Applicants: Panasonic Intellectual Property Management Co., Ltd.
- Inventors: Masatoshi Matsuo; Tsuyoshi Nakamura; Tadanori Tezuka
- Web page: https://patentgrep.com/patent/US10441198B2

## CPC

- A61B 5/1171: based on the shapes or appearances of their bodies or parts thereof
- A61B 5/02416: {using photoplethysmograph signals, e.g. generated by infrared radiation (A61B5/14552 takes precedence)}
- A61B 5/742: {using visual displays (displays for heart-related electrical signals, e.g. ECG, A61B5/339)}
- G06V 10/56: relating to colour
- G06V 40/161: {Detection; Localisation; Normalisation}
- G06V 40/172: {Classification, e.g. identification}
- G06V 40/40: Spoof detection, e.g. liveness detection
- A61B 2576/02: specially adapted for a particular organ or body part
- G06F 2218/12: Classification; Matching
- G06V 40/15: {Biometric patterns based on physiological signals, e.g. heartbeat, blood flow}
- G16H 30/40: for processing medical images, e.g. editing
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

Face detection device that detects a face from an image including a person. The device includes a processor and a memory including instructions that, when executed by the processor, cause the processor to perform operations including: receiving captured images by a camera, detecting one face region or more in the captured images, extracting vital information which includes an index indicating a living state of a face imaged in the face region, based on image data regarding the detected face region; and determining whether or not the face in the face region is in a living state, based on the vital information. The face region detector recognizes a face region corresponding to the face which is determined to be in the living state by the vitality determiner among the one face region or more, as a face region based on a living body.

## Claims

### Claim 1

A face detection device that detects a face from an image including a person, the device comprising:
a processor; and
a memory including instructions that, when executed by the processor, cause the processor to perform operations including:
receiving captured images by a camera;
detecting at least one face region in the captured images;
collating a feature quantity of a face in a face region with information of a face which is previously registered;
determining whether a vitality detection mode has been selected, the vitality detection mode configured to recognize a face of a person in a living state but not a face of an object in a non-living state;
when the vitality detection mode is determined to have been selected:
extracting vital information which includes an index indicating the living state of a face imaged in the at least one face region, based on image data regarding the detected at least one face region;
determining whether or not the face in the at least one face region is in the living state, based on the vital information; and
when the face in the at least one face region is determined to be in the living state, recognizing a face region corresponding to the face which is determined to be in the living state among the at least one face region, as a face region of a living body, and the collating being performed by using of a feature quantity of the face in the face region corresponding to the face which is determined to be in the living state.

### Claim 2 (depends on 1)

The face detection device of claim 1,
the instructions further include estimating a pulse in the face imaged in the face region, based on the image data regarding the at least one face region, and extracts the estimated pulse as the vital information.

### Claim 3 (depends on 1)

The face detection device of claim 1, further comprising:
a display configured to display the captured images, and
the instructions further include performing image processing for causing a user to distinguish a face region based on the living body from another face region among the at least one face region, on the captured images displayed on the display.

### Claim 4 (depends on 1)

The face detection device of claim 1,
wherein the collating is applied only on the face which is determined to be in the living state in a case where the face is determined to be in the living state.

### Claim 5 (depends on 1)

The face detection device of claim 1, further comprising:
determining a face which is determined to be not in the living state, to be invalid or discrepant.

### Claim 6 (depends on 1)

A face detection system comprising:
the detection device of claim 1; and
the camera configured to input the captured images to the face detection device.

### Claim 7 (depends on 1)

The face detection device of claim 1, further comprising:
when the face in the at least one face region is determined not to be in the living state, not performing recognizing of a face region corresponding to the face which is determined not to be in the living state among the at least one face region.

### Claim 8 (depends on 1)

The face detection device of claim 1,
wherein, when the vitality detection mode is not selected, recognizing a face region corresponding to the face which is determined not to be in the living state,
wherein the recognizing is performed on each of the at least one face region detected, and
wherein the at least one face region detected includes a face of a non-living object.

### Claim 9

A face detection method of detecting a face from a captured image including a person, the method comprising:
receiving an input of captured images which include the person and are continuous in time;
detecting at least one face region in the captured images;
collating a feature quantity of a face in a face region with information of a face which is previously registered;
determining whether a vitality detection mode has been selected, the vitality detection mode configured to recognize a face of a person in a living state but not a face of an object in a non-living state;
when the vitality detection mode is determined to have been selected:
extracting vital information which includes an index indicating that a face imaged in the at least one face region is in the living state, based on image data regarding the detected at least one face region;
determining whether or not the face in the at least one face region is in the living state, based on the vital information; and
when the face in the at least one face region is determined to be in the living state, recognizing a face region corresponding to the face which is determined to be in the living state among the at least one face region is recognized as a face region of the a living body, and the collating being performed by using of a feature quantity of the face in the face region corresponding to the face which is determined to be in the living state.