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

A Fast and Robust Face Recognition Method

A face recognition and robust person technology, applied in the field of fast and robust face recognition, can solve problems such as robustness, and achieve the effect of improving robustness and computing efficiency

Active Publication Date: 2021-08-31
ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF0 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 the robustness of the classic deep neural network, this invention proposes a new fast and robust face recognition method, which has good robustness to occlusion, posture and other factors

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
  • A Fast and Robust Face Recognition Method
  • A Fast and Robust Face Recognition Method
  • A Fast and Robust Face Recognition Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0080] The present invention will be further described below in conjunction with specific examples, but the protection scope of the present invention is not limited thereto.

[0081] refer to figure 1 , a fast and robust face recognition method. The face recognition process is divided into two parts. The first part is to obtain the probability of the test picture in each category through the deep neural network, which is denoted as p ic , i=1,2,...,c, c represents the category of the picture; the second block is to obtain the probability of the test picture through sparse weighted representation, denoted as p ir , i=1,2,...,c; through the weight ω c and ω r to calculate the test picture probability P i =p ic ×ω c +p ir ×ω r , i=1,2,…,c; the result of the final classification is .

[0082] It specifically includes the following steps:

[0083] S1. Assuming that there are c classes in the training sample set, each training sample X ik , i=1,2,...,c, k=1,2,...n i ...

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 relates to a fast and robust face recognition method, which distributes the weights of the deep neural network and the weighted sparse representation through the error of the weighted sparse representation, and integrates the prediction value of the deep neural network and the weighted sparse representation according to the weight to improve the robustness. Stickiness, in order to improve the computational efficiency of weighted sparse representation, the image is also subjected to sparse low-rank decomposition to obtain the subspace sample matrix and residual matrix, and the residual matrix is ​​processed into a vector as the bias initialization of weighted sparse representation can make the convergence speed faster Faster, the present invention has better robustness than the traditional deep neural network for face recognition, and the recognition rate is not much different from that of the deep neural network in the absence of noise, and inherits the accuracy of the deep neural network in the absence of noise. It is more robust than deep neural networks in the presence of noise.

Description

technical field [0001] The invention belongs to the field of new generation information technology, and in particular relates to a novel fast and robust face recognition method. Background technique [0002] With the development of artificial intelligence and computer technology, face recognition is widely used in various fields, such as face payment, face entry, ID card inspection, etc. With the continuous development of society, face recognition will be applied to more many places. [0003] At this stage, with the continuous development of the neural network, the face recognition rate can reach a high level through the training of the neural network, but in the actual situation, it will be restricted by various factors, and the data that is different from the training distribution will be evaluated. When the performance is not very good, such as face occlusion, illumination, posture, etc., these factors will affect the recognition rate of the face. [0004] In order to s...

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): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V40/16G06V40/172G06F18/2136G06F18/214
Inventor 于爱华唐明侯北平朱文李刚朱广信杨舒捷朱必宏宣仲伟宣皓莹
Owner ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com