Unlock instant, AI-driven research and patent intelligence for your innovation.

Peephole system based on face recognition and face recognition method thereof

A face recognition and door mirror technology, which is applied in the field of face recognition, can solve the problems of improving the robustness of video face recognition, not effectively filtering and compensating for face changes, and achieving fast calculation speed and improved recognition efficiency. Effect

Inactive Publication Date: 2017-11-03
SUZHOU TIANPING ADVANCED DIGITAL TECH
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] One method of traditional video to static image recognition is to use static face recognition for each frame of face image to identify, and finally, according to probability, distance judgment, etc., to synthesize all recognition results; another method is to use video Sequential generation of face models or methods of generating face models based on motion timing information; the above two methods do not effectively filter and compensate for various face changes in videos, so the robustness of video face recognition needs to be improved

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Peephole system based on face recognition and face recognition method thereof
  • Peephole system based on face recognition and face recognition method thereof
  • Peephole system based on face recognition and face recognition method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] This embodiment discloses a door mirror system based on face recognition. The door mirror system is mainly used in smart home systems. Face detection and face recognition functions are added to the outdoor unit of the smart home. When the sensor of the outdoor unit detects When someone visits, the video stream is obtained through the camera of the outdoor unit, and continuous face detection is performed on it.

[0037] Specifically, such as figure 1As shown in , the door mirror system mainly includes: entrance light, outdoor unit, indoor unit, cloud server, and mobile terminal. The above outdoor unit is a face recognition device, and the above outdoor unit is installed at the entrance of the door mirror system. The machine is installed indoors and interacts with the information of the outdoor unit. The outdoor unit acquires face images, detects and recognizes them, and then outputs the recognition result to the indoor unit. The information obtained by the indoor unit is...

Embodiment 2

[0051] A kind of face recognition method is disclosed in embodiment 2, comprises the following steps:

[0052] The sensor of the S 101 outdoor unit detects that someone is approaching, and the video stream of the face image is obtained through the camera;

[0053] S 102 Outdoor Unit According to the feature value of the face, the trained classifier traverses each pixel of the input face image with different scales during the detection process of the face detection module to detect faces of different sizes. The similarity between the obtained face and the face in the face database is higher than the threshold, and it is identified whether it is a member registered in the database;

[0054] S 103 transmits the recognition result to the indoor unit;

[0055] S 104 When the indoor unit does not respond, the indoor unit sends the received face image to the cloud server, and the cloud server sends it to the remote mobile terminal.

[0056] In the above-mentioned embodiment, based ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a peephole system based on face recognition. The system comprises a home-entry lamp, an outdoor machine, an indoor machine, a cloud server and a mobile terminal. The outdoor acquires a face image, carries out detection and recognition, and outputs a recognition result to the indoor machine. The indoor machine transmits the acquired information to the cloud server. The cloud server transmits the information to the mobile terminal. The home-entry lamp is controlled by the outdoor machine. When the outdoor machine detects a face, the home-entry lamp is automatically turned on. The outdoor machine comprises an input image module, a face detection module, a characteristic extraction module, a face recognition module and a result outputting module. According to the invention, by use of the statistical method, modeling is performed on the face and by comparing the matching degree of all possible to-be-detected regions with a face model, a possibly existing face region is obtained; and through the method based on a Haar classifier, the computation speed is quick and the recognition efficiency can be improved.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a door mirror system based on face recognition and a face recognition method thereof. Background technique [0002] With the development of video surveillance technology, face recognition based on video has achieved good development, and it plays a very important role in security and security fields such as intelligent transportation, access control, and information security. The existing video face recognition method is a kind of video recognition of static images, using face video as input, and realizing recognition or verification by comparing with the face database of static images. [0003] One method of traditional video to static image recognition is to use static face recognition for each frame of face image to identify, and finally, according to probability, distance judgment, etc., to synthesize all recognition results; another method is to use video Sequentia...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00H04N7/18E06B7/30
CPCH04N7/186E06B7/30G06V40/172G06V40/161G06V40/168
Inventor 许伟刚
Owner SUZHOU TIANPING ADVANCED DIGITAL TECH