Check patentability & draft patents in minutes with Patsnap Eureka AI!

A Face Recognition Method with Multiple Sparse Representations in Small Samples

A sparse representation, face recognition technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve complex cost and other problems, and achieve the effect of strong robustness and good classification effect

Inactive Publication Date: 2017-10-17
EAST CHINA JIAOTONG UNIVERSITY
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the solution of this kernel sparse coding model is more complicated and expensive than the classical sparse representation model

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 Face Recognition Method with Multiple Sparse Representations in Small Samples
  • A Face Recognition Method with Multiple Sparse Representations in Small Samples
  • A Face Recognition Method with Multiple Sparse Representations in Small Samples

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The present invention will be further described now in conjunction with accompanying drawing, see figure 1 , a multi-sparse representation classification method, including the following specific steps:

[0035] (1) Input sample 101 and produce virtual training sample 102; In this process, the face image is stored in matrix form, and the size of the matrix is ​​long and high all set to even numbers, to facilitate the follow-up mirror transformation operation. Two mirror operations are used to generate virtual training samples, and the specific process of one mirror transformation is as follows: record any image matrix as I, and its mirror image matrix as M, then M(i,j)=I(i,t-j+1 ), i=1, 2,..., s, j=1, 2,..., t, where s and t are the number of rows and columns of the image I, respectively.

[0036] (2) The feature extraction process includes three methods KPCA, KDA and KLPP, and their corresponding processes are 103, 104 and 105 respectively. In this process, face image...

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

A face recognition method with multi-sparse representation in the case of small samples. This method uses two methods to solve the small sample situation in face recognition. One is to generate "virtual samples" from the given original training samples, and increase the training samples Second, on the basis of generating virtual samples, use three nonlinear feature extraction methods, namely kernel principal component analysis, kernel discriminant analysis and kernel local preservation projection algorithm, to extract the characteristics of samples; in this way, three types of feature patterns will be obtained , construct a sparse representation model for each feature pattern; construct a total of three sparse representation models for each sample, and finally classify according to the representation results. The multi-sparse representation classification method provided by the present invention generates a virtual human face through symmetrical mirroring, and then constructs and classifies a multi-sparse representation model based on the L1 norm. Compared with other classification methods, this method has strong robustness and good classification effect, and is especially suitable for many classification occasions with high data dimension and few training samples.

Description

technical field [0001] The invention relates to a face recognition method with multi-sparse representation in the case of small samples, and belongs to the technical field of pattern recognition and machine learning. Background technique [0002] With the development of technologies such as computers, networks, and multimedia, people need to process more and more high-dimensional and complex data such as images and videos, and most of the processing of these data is classification or identification. In recent years, an important branch of image recognition, that is, biometric recognition, is in the ascendant, and it is a research hotspot in the field of pattern recognition. Compared with other biometric identification technologies such as fingerprint identification, face recognition has been widely concerned and used due to its ease of use. For example, after 9 / 11, the United States adopted face recognition systems in several airports. Both the 2008 Beijing Olympic Games an...

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): G06K9/66G06K9/46
CPCG06V40/168G06F18/24G06F18/214
Inventor 范自柱倪明康利攀
Owner EAST CHINA JIAOTONG UNIVERSITY
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