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

Method and device for single sample face identification based on local convolution feature combination representation

A convolution and partial technology, applied in the field of computer vision and pattern recognition, can solve the problems of low recognition efficiency, poor robustness, and low recognition accuracy, so as to reduce time consumption, improve robustness and discrimination ability, and improve efficiency and the effect on recognition accuracy

Inactive Publication Date: 2017-11-07
SHENZHEN UNIV
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a single-sample face recognition method and device based on the joint representation of local convolution features

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
  • Method and device for single sample face identification based on local convolution feature combination representation
  • Method and device for single sample face identification based on local convolution feature combination representation
  • Method and device for single sample face identification based on local convolution feature combination representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] figure 1 It shows the implementation process of the single-sample face recognition method based on the joint representation of local convolution features provided by Embodiment 1 of the present invention. For the convenience of explanation, only the parts related to the embodiment of the present invention are shown, and the details are as follows:

[0025] In step S101, a face image to be recognized is received, and corresponding image blocks to be recognized are extracted from a preset number of feature points of the face image to be recognized by a preset feature point division method.

[0026] In the embodiment of the present invention, the feature point division method is used to extract corresponding image blocks to be recognized at the preset number of feature points of the face image to be recognized, that is, the face image to be recognized is divided into a preset number of images to be recognized piece.

[0027] Specifically, the key feature points of the fac...

Embodiment 2

[0042] figure 2 It shows the flow of local adaptive convolution network training, intra-class change dictionary, query dictionary, projection matrix and temporary matrix calculation process in the single-sample face recognition method based on local convolution feature joint representation provided by Embodiment 2 of the present invention , for ease of description, only the parts related to the embodiment of the present invention are shown.

[0043] In the embodiment of the present invention, the training process in the face recognition method trains to obtain a local adaptive convolutional network, and calculates the intra-class change dictionary, query dictionary, projection matrix and temporary matrix, as follows:

[0044] In step S201, extract the corresponding first image block at each feature point on the preset first training library face image, and train the local adaptive convolutional network corresponding to each feature point according to each first image block ....

Embodiment 3

[0060] image 3 The structure of the single-sample face recognition device based on the joint representation of local convolution features provided by Embodiment 3 of the present invention is shown. For the convenience of description, only the parts related to the embodiment of the present invention are shown, including:

[0061] The image division unit 31 is used to receive the face image to be recognized, and extract corresponding image blocks to be recognized at a preset number of feature points of the face image to be recognized by a preset feature point division method;

[0062] The feature extraction unit 32 is used to input each image block to be identified into a well-trained local adaptive convolution network corresponding to each feature point, so as to extract the feature of each feature point on the face image to be identified;

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 present invention is suitable for the computer vision and mode identification technology field, and provides a method and device for single sample face identification based on local convolution feature combination representation. The method comprises: extracting corresponding image blocks to be identified at feature points with preset quantity of a face image to be identified in a feature point division mode; putting each identification image to be identified into a local adaptation convolution network corresponding to each feature point to extract the features of each feature point; calculating combination representation of the features of all the feature points on the face image to be identified according to a changing dictionary and a query dictionary in a category, and calculating the representation coefficient of the features of all the feature points in the combination representation according to a projection matrix and a temporary matrix; and according to the representation coefficient, the changing dictionary in the category and the features of each feature point on the face image to be identified, determining the identity of the face image to be identified so as to effectively improve the face identification robustness, reduce the face identification time consumption and effectively improve the face identification efficiency and identification accuracy.

Description

technical field [0001] The invention belongs to the technical field of computer vision and pattern recognition, and in particular relates to a single-sample face recognition method and device based on joint representation of local convolution features. Background technique [0002] In the field of computer vision and pattern recognition, face recognition has huge market value and has been widely concerned by academia and industry. In the process of face recognition, there are usually large differences in the face images taken by the same person in different environments, such as partial occlusion of the face, different lighting, different expressions, and different postures. Therefore, the face recognition algorithm is required to have better robustness. In practical applications, each person in the database may only have one face image (such as an electronic passport face image, a driver's license face image, etc.), and face recognition in this case is called single-sample...

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): G06K9/00G06K9/62
CPCG06V40/168G06F18/214
Inventor 杨猛王兴
Owner SHENZHEN 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