Human-face identifying method and device

A face recognition and face image technology, applied in the field of recognition, can solve the problems of large number of training samples, computational complexity and difficulty of similarity learning, etc., to improve accuracy, reduce computational complexity, and reduce data volume. Effect

Active Publication Date: 2014-03-26
四川哈工创兴大数据有限公司
View PDF2 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In view of this, the purpose of the present invention is to provide a method and device for face recognition to solve the prob

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
  • Human-face identifying method and device
  • Human-face identifying method and device
  • Human-face identifying method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] see figure 1 , which shows a flow chart of Embodiment 1 of a face recognition method provided by the present application, including:

[0057] Step S101: performing a first classification process on the face image samples to obtain a training sample group and a test sample group respectively;

[0058] In the process of using a class of support vector machines to recognize face images, the class of support vector machines is firstly trained.

[0059] The face image samples in the database are divided into two categories, including: training sample group and test sample group.

[0060] The training sample group is used to train a class of support vector machines to make its parameters more precise and more accurate. The test sample group is used to test the one-class support vector machine after the training, so as to detect the recognition accuracy of the one-class support vector machine after the training.

[0061] Step S102: Perform a second classification process on...

Embodiment 2

[0077] see figure 2 , showing a flow chart of Embodiment 2 of a face recognition method provided by the present application, figure 1 In the flow chart shown, step S102 includes:

[0078] Step S1021: Divide the training samples in the training sample group into at least two subsets according to preset classification conditions, and each subset corresponds to one type of data;

[0079] According to the classification condition, the classification condition may be various conditions related to face distinction such as age, gender, and race.

[0080] The training samples in the training sample group are divided into multiple subsets, and each subset is a class of data.

[0081] It is assumed that there is a set of face training samples, which are displayed in the form of data, {(x 1 ,v 1 ),…,(x i ,v i ),…,(x n ,v n )}, where x i ∈ R D , v i ∈{1,2,...,C}.

[0082] v i is x i The category label for the class indicating the category of the classification. In this emb...

Embodiment 3

[0096] see image 3, showing a flow chart of Embodiment 3 of a face recognition method provided by the present application, figure 2 In the flow chart shown, step S103 includes:

[0097] Step S1031: select the kernel function as the Gaussian radial basis function, and preset kernel parameter values;

[0098] The Gaussian radial basis function is selected as the kernel function of a class of support vector machine, and the Gaussian radial basis function k(z, z′)=e -σ||z-z′|| , where the σ is a kernel parameter, and the value of the kernel parameter is preset according to the actual situation.

[0099] Step S1032: input the training sample pair into the kernel function, train a type of support vector machine, and obtain model coefficients;

[0100] After specifying the experimental parameters, use the Gaussian radial basis function as the kernel function, input the training samples in the training sample pair group, train this type of execution vector machine, and obtain the...

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 provides a human-face identifying method. The method is used for learning the similarity among human-face images on the basis of a one-class support vector machine and comprises the following steps of classifying human-face samples to obtain a training-sample set and a testing-sample set; classifying training samples in the training-sample set to obtain at least two classes, acquiring the training samples in each class to generate difference-sample pairs, and constructing a training-sample pair group; training the one-class support vector machine according to the training-sample pair group to obtain the decision-making model parameter of the one-class support vector machine and obtain a similarity discriminating model; carrying out similarity judgment by inputting testing difference samples generated by two testing samples which are randomly acquired in the testing-sample set into the similarity discriminating model. In the method, the data quantity input into the one-class support vector machine is reduced by classifying the training samples input into the one-class support vector machine according to the mode of generating training-sample differences in the homogeneous training samples, so that the calculating complexity is lowered.

Description

technical field [0001] The invention belongs to the technical field of recognition, and in particular relates to a face recognition method and device. Background technique [0002] An information-rich pattern collection of human faces is the main symbol for human beings to distinguish, recognize and remember each other. Face recognition plays an important role in computer vision, pattern recognition, and multimedia technology research, so face recognition technology is one of the most challenging research topics in the field of pattern recognition and computer vision. [0003] The main work of face recognition is to map the face image in the real space to the machine space, and to describe the face as completely and accurately as possible in a certain way (such as the geometric features, algebraic features and transformation coefficients of the face) . Compare the face to be recognized with the known face, and judge the identity of the face according to the degree of simil...

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
Inventor 张莉卢星凝曹晋王邦军何书萍李凡长杨季文
Owner 四川哈工创兴大数据有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap