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

Semi-supervised learning face recognition method based on lrr graph

A semi-supervised learning and face recognition technology, which is applied in the field of semi-supervised learning face recognition based on low-rank representation graphs, can solve the problems of reducing the accuracy of face recognition and achieve strong robustness

Inactive Publication Date: 2016-03-02
XIDIAN UNIV
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 since the method of the patent application is a supervised face recognition method, the accuracy of face recognition is reduced when the number of known samples and corresponding labels is small, which leads to the fact that the method is not effective in the case of small sample learning. Reduced face recognition accuracy due to lack of supervisory information

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
  • Semi-supervised learning face recognition method based on lrr graph
  • Semi-supervised learning face recognition method based on lrr graph
  • Semi-supervised learning face recognition method based on lrr graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] refer to figure 1 , the present invention is described in further detail.

[0029] Step 1, divide the database sample set. Take 3 samples from each class in the Yale face database as the training set A∈R D×M , and the remaining samples are used as the test set B∈R D×T .

[0030] Among them, D represents the dimensionality of training set samples and test set samples, R n Represent n-dimensional real number space, M is the total number of training set samples, and T is the total number of test set samples; in an implementation example of the present invention, the sample dimension D is 8000, the total number M of training set samples is 45, and the number of test set samples The total T is 125.

[0031] Step 2, form a sample set.

[0032] 2a) Arrange the samples in the training set before the test set samples in order of labels to form the original sample matrix;

[0033] 2b) The random matrix Q∈R d×D Multiply by the original sample matrix to get the dimensionall...

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 semi-supervised learning face recognition method based on a low-rank representation (LRR) graph. The method comprises the following steps of: (1) dividing sample sets in a database; (2) forming a sample set; (3) generating an initial label matrix; (4) performing LRR; (5) generating a sample similarity matrix; (6) generating a class probability matrix; and (7) outputting the class of a test sample. By the semi-supervised learning method, high recognition correct rate can be achieved under the condition of a few known label samples; and meanwhile, by the LRR method, high robustness is achieved under the condition that the sample is subjected to noise pollution.

Description

technical field [0001] The present invention belongs to the technical field of image processing, and further relates to a face recognition method in the technical field of pattern recognition, specifically a semi-supervised learning face recognition method based on a Low-Rank Representation (LRR) graph, which can be used in the present invention Identity recognition in video surveillance environment, network security and security fields. Background technique [0002] Face recognition is different from iris recognition, fingerprint recognition, etc. It is a non-invasive recognition method that is relatively easy to be accepted by people, and thus becomes a technology that needs a breakthrough in the field of computer vision and pattern recognition technology. The main task of face recognition is to compare the extracted face features with the face images in the database to determine the identity of the face to be recognized. At present, methods for recognizing arbitrary face...

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): G06K9/62G06K9/00
Inventor 杨淑媛焦李成王秀秀刘芳缑水平侯彪王爽马文萍杨丽霞徐雯辉谢冬梅
Owner XIDIAN UNIV