Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Living body detection method based on facial microexpression recognition

A technology of liveness detection and micro-expression, which is applied in the field of artificial intelligence face recognition and analysis, which can solve the problems of high privacy of data verification, easy leakage of face portraits, and leakage of face information, and achieve simple and fast verification methods , highly operable effect

Inactive Publication Date: 2018-12-18
TIANJIN SEWEILANSI TECH +1
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But while face recognition brings convenience, there are many hidden dangers
Face portraits are easily leaked in daily life. If the data verification bound to face verification is highly private, it will cause great losses to users.
Therefore, how to prevent the leakage of face information, and the problem of invalid verification has become a crucial problem in face recognition technology.

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
  • Living body detection method based on facial microexpression recognition
  • Living body detection method based on facial microexpression recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] 1) Use opencv to access the local or network camera;

[0040] 2) Use the DLIB framework face detection method to detect whether the face appears in the video, and cut and store the face area to obtain face face1;

[0041] 3) The system randomly generates an expression, and guides the verification to make an expression;

[0042] 4) Pass the face area face1 into the model trained with the caffe deep learning network, identify the facial expression label1, and judge whether it is consistent with the required facial expression: if yes, continue to the next verification; if not, you need to start again verify;

[0043] 5) Execute the above steps 2) and 3) again to obtain face2;

[0044] 6) Use DLIB convolutional network face recognition to judge the similarity between face1 and face2. If the similarity exceeds a predetermined threshold, it is considered to be a person.

Embodiment 2

[0046] Train facial expression recognition with the deep learning framework TENSORFLOW. Because TENSORFLOW is a framework developed based on the python language, for compatibility, other frameworks must also use the version corresponding to the python interface during implementation.

[0047] 1) Use opencv_python to access the local or network camera;

[0048] 2) Use python_dlib to detect and crop living people;

[0049] 3) Pass the user's micro-expression into the network trained by tensorflow, and obtain the expression result predicted by the tensorflow framework;

[0050] 4) After the expression is successfully verified once, the living person is detected again and cropped;

[0051] 5) Use python_dlib convolutional network face recognition to judge whether the two results are a person.

[0052] The tensorflow main application is deployed in the network, which can be a cloud server or a LAN server.

Embodiment 3

[0054] Torch is mainly a deep learning training framework under the Lua language architecture.

[0055] Using the Torch method is similar to the steps of tensorflow, except that the model trained by the deep learning network in step 3) is Torch, and the face detection, positioning, and recognition algorithms can only be redefined and implemented using the algorithms provided by the framework of Torch itself. In terms of use, Torch has poor compatibility with the current mainstream ready-made image recognition library openCVDlib, and there are few materials. Therefore, when using the torch framework, the interface of the torch framework is used. Since the number of people used by Torch is relatively small compared with caffe and tensorflow, it is difficult to detect people. There is a certain gap in the accuracy and efficiency of the face.

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 belongs to the technical field of artificial intelligence face recognition and analysis application, and relates to a technology for face recognition application belong to data securityverification technology, in particular to a living body detection method based on facial microexpression recognition. At first, a computer trains an N-type expression through a neural network in advance, and then during log-in for authentication, the computer prompts the user to make M consecutive expressions, and the user needs to follow the computer to make M consecutive expressions; secondly, through judgment by a neural network, K*M pictures are captured during the period of making expressions, where K>= 2, whether K*M faces are the same person is judged, finally Q faces are taken out of K*M pictures, where Q> = 1, and the average result obtained after many times verification is taken to verify the face identity.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence face recognition and analysis application, relates to data security verification technology in face recognition application, and specifically relates to a living body detection method based on face micro-expression recognition. Background technique [0002] As the most popular artificial intelligence landing technology, face recognition has been widely used in many fields. But while face recognition brings convenience, there are many hidden dangers. Face portraits are easily leaked in daily life. If the data verification bound to face verification is highly private, it will cause great losses to users. Therefore, how to prevent the leakage of face information, and the problem of verification failure caused by it has become a crucial problem in face recognition technology. This technique has become known in the professional field as human biopsy. At present, there are mai...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00
CPCG06V40/174G06V40/40
Inventor 王光夫赵洁孙杰楼嘉昌
Owner TIANJIN SEWEILANSI TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com