Target feature selection method based on three-way decision

A target feature and decision-making technology, which is applied in the field of target recognition to achieve the effect of improving the fault tolerance rate, recognition performance and solving target recognition problems.

Pending Publication Date: 2021-08-10
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing classical rough set theory has defects in dealing with uncertain data and numerica

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
  • Target feature selection method based on three-way decision
  • Target feature selection method based on three-way decision
  • Target feature selection method based on three-way decision

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0064] Select four types of samples from the remote sensing image set NWPU-RESISC45 Dataset, including beaches, forests, highways, and islands. There are 12 samples in each type and a total of 48 samples. The remote sensing images of each type are as follows: figure 2 shown.

[0065] A total of 24 features of color features and texture features are extracted from all pictures, and 24 features are selected.

[0066] 1. Use the ReliefF algorithm to get the weight value of all features (a total of 24 features) W={W 1 ,W 2 ,...,W 24}.

[0067] A four-category problem C={c 1 ,c 2 ,c 3 ,c 4} sample set S={s 1 ,s 2 ,...,s 48}, each sample contains 24 features, namely s p ={s p(1) ,s p(2) ,...,s p(24)}, 1≤p≤48, the values ​​of all features are numeric, then define two samples s i , s j The distance on feature g is:

[0068]

[0069] 1.1 First initialize all feature weight sets

[0070] 1.2 Randomly select a sample s from the training sample set S, and then find...

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 target feature selection method based on a three-way decision. A feature selection algorithm based on a three-way decision theory is used for solving the identification problem under a high-dimension small sample. Aiming at the limitation that only one threshold value in a typical filtering algorithm ReliefF is used as a feature accepting and rejecting condition and the defect that a packaging algorithm needs a large amount of execution time, three decisions are introduced, the filtering algorithm and the packaging algorithm are combined, one threshold value is expanded into two threshold values on the basis of the traditional ReliefF algorithm, and the features are split into a positive domain, a negative domain and a boundary domain according to the feature weights; and the features of the three domains are selected respectively, so that the error-tolerant rate of the algorithm is increased to a certain extent, and the recognition performance is greatly improved. According to the invention, the accuracy of the learning model is used as a selection standard, and the defect of other algorithms in recognition accuracy is overcome.

Description

technical field [0001] The invention belongs to the technical field of target recognition, and in particular relates to a target feature selection method. Background technique [0002] With the rapid development of information technology, various fields have ushered in the era of big data. Big data includes two aspects: one is the large number of samples in the data set; the other is that the data contains large dimensions. With the advent of the era of big data, data mining has ushered in a wave of research, and object recognition is a kind of data mining. At present, more and more research results have been produced for massive and complex images, texts and other data. However, there may not necessarily be a large number of labeled samples in practical applications, for example, remote sensing pictures in military fields such as aerospace. In the case of high-dimensional data, most of the existing traditional algorithms are suitable for a large number of labeled samples...

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): G06K9/62
CPCG06F18/217G06F18/2411G06F18/214
Inventor 李波骆双双田琳宇万开方高晓光
Owner NORTHWESTERN POLYTECHNICAL UNIV
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