Pattern classification method based on partial linear representation

A pattern classification and local linear technology, applied in the field of pattern recognition, can solve the problem of high time complexity of sparse coefficients, achieve the effect of reducing the number of training samples, reducing the difficulty of calculation, and improving the recognition rate

Inactive Publication Date: 2013-08-14
NANJING UNIV OF INFORMATION SCI & TECH
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the biggest problem with the SRC classifier is that the time complexity of calculating sparse coeffic

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
  • Pattern classification method based on partial linear representation
  • Pattern classification method based on partial linear representation

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0019] The technical scheme of the invention is described in detail below:

[0020] Using the training sample set X containing c categories to identify the category to which the test sample y belongs, includes the following steps.

[0021] X=[X 1 ,X 2 ,...,X c ], X i =[x i1 , X i2 ,...,X iNi ] Represents the i-th training sample set, X i Contains N i Samples, x ij ∈R d (R d Represents the d-dimensional real vector set) represents the j-th training sample of the i-th category, y∈R d , C is a natural number, N i Is a natural number.

[0022] Step 1. For the test sample y, construct its neighbor training sample set:

[0023] Step 1-1, calculate the test sample y to each training sample x in the training sample set X ij Distance D ij .

[0024] D ij =||y-x ij ||(i=1,2,…,c,j=1,2,…,N i );

[0025] Step 1-2, extract the first K nearest neighbor training samples x from the training sample set 1 ,x 2 ,...,X K , Forming the nearest neighbor training sample set X′=[x 1 ,x 2 ,...,X K ], w...

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 pattern classification method based on partial linear representation, and belongs to the technical field of pattern recognition. The method comprises the steps as follows: firstly, test samples are represented by using partial neighbor training samples of the test samples, so that a group of linear representation coefficients are obtained; then reconstruction errors of the test samples are reconstructed by using each type of samples in the neighbor training samples and corresponding linear representation coefficients; and finally, the test samples are classified according to the reconstruction errors. When the linear representation of the test sample is calculated, the number of the training samples and the calculation difficulty are reduced; and the recognition rate is increased, and the calculation time is shortened at the same time.

Description

technical field [0001] The invention discloses a pattern classification method based on local linear representation, and belongs to the technical field of pattern recognition. Background technique [0002] SRC (Sparse Representation-based Classification, sparse representation classifier), using all samples as a dictionary, sparsely represents the relationship between test samples and training samples. Compared with the traditional nearest neighbor classifier, the SRC classifier has achieved better classification performance. However, the biggest problem with the SRC classifier is that the time complexity of calculating sparse coefficients is very high. As the number of training samples increases, the calculation time increases exponentially. Contents of the invention [0003] The technical problem to be solved by the present invention is to provide a pattern classification method based on local linear representations in view of the above-mentioned deficiencies in the back...

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): G06K9/66
Inventor 刘茜马杰良王丽娜
Owner NANJING UNIV OF INFORMATION SCI & TECH
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