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

Face Recognition Method Based on Mask

A face recognition and mask technology, applied in the field of mask-based face recognition with occlusion, can solve the problems of uncertain face occlusion and high false recognition rate, achieve effective classification processing, improve recognition rate, and reduce interference The effect of features

Active Publication Date: 2019-03-22
SHANGHAI JIAOTONG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of high misrecognition rate caused by uncertain face occlusion mentioned in the above content, the present invention provides a face recognition method with occlusion based on masks for different occlusion scenarios

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
  • Face Recognition Method Based on Mask
  • Face Recognition Method Based on Mask
  • Face Recognition Method Based on Mask

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0043] Such as figure 1 Shown, is the general flow chart of the method of an embodiment of the present invention:

[0044] The first step is the training process. According to the sample data, train the PLDA face classification model MDL. The specific steps include:

[0045] 1.1) Convert the original color image into a grayscale image I(x,y);

[0046] 1.2) Use the face detection algorithm based on the Viola and Jones framework and the face alignment method based on SDM and block radial transformation for the image I(x,y) to detect and align the face, and obtain the aligned face area image In this example, It is an image whose length and width are 128×128;

[0047] 1.3) For face images Add the corresponding mask {M} respectively n , get the picture where n represents the type of mask. right i=1,2,3...n. mask{M(x,y)} n is a binary image of 0 or 255, n=2 in this embodiment, where {M(x,y)} 1 Masks for glasses occlusion, such as figure 2 Shown; {M(x,y)} 2 Masks f...

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 mask-based occluded face recognition method, which includes the following steps: the first step, the training process, according to the sample data, training the PLDA face classification model MDL; the second step, the training process, according to the sample data , train the occlusion detection model; the third step, the recognition process, detect the occlusion type of the data to be recognized according to the occlusion detection model, and perform face recognition according to the corresponding PLDA face classification model MDL. The invention utilizes a human face occlusion detection algorithm, uses a mask to cover the occluder, and forms a new human face subspace. For different occlusions, PLDA classifiers are trained in different face subspaces, and finally the similarity of sample pairs is calculated by face comparison. The invention is a method capable of improving the face recognition rate in the case of occlusion, and fully utilizes different face subspaces by adding masks to improve the face recognition accuracy rate.

Description

technical field [0001] The present invention relates to an occlusion recognition method in the technical field of face recognition, in particular to a mask-based face recognition method with occlusion. Background technique [0002] In the actual process of face image processing, face image occlusion (such as glasses, hats, etc.) often occurs, and occlusion has a great impact on the performance of face recognition. Occluded face images are divided into unintentional occlusion and intentional occlusion. Common unintentional occlusions include hats, glasses, etc., while intentional occlusions are usually sunglasses, masks, or other objects that block facial features. Intentional occlusion is usually due to excessive changes in features, which can easily cause recognition failures and lead to high misidentifications. Unintentional occlusion usually only occludes a small part of facial features, which easily leads to the introduction of too many disturbing features in the featu...

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 Patents(China)
IPC IPC(8): G06K9/00
CPCG06V40/168G06V40/172
Inventor 杨小康潘岑蕙晏轶超徐奕
Owner SHANGHAI JIAOTONG UNIV
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