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

Image dimension reduction clustering method based on fuzzy theory

A clustering method and fuzzy theory technology, applied in the field of machine learning, can solve the problems of losing category information, reducing clustering accuracy, reducing algorithm efficiency, etc., to achieve the goal of speeding up computing time, improving algorithm efficiency, and reducing computational complexity Effect

Pending Publication Date: 2022-08-05
NORTHWESTERN POLYTECHNICAL UNIV
View PDF2 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The methods in these production practices are to reduce the dimensionality and cluster the data step by step. The algorithm is divided into two steps to reduce the efficiency of the algorithm, and the category information may be lost in the process of dimensionality reduction, reducing the clustering accuracy, so it can be considered Combine dimensionality reduction and clustering into one method to improve algorithm efficiency and reduce loss of image category 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
  • Image dimension reduction clustering method based on fuzzy theory
  • Image dimension reduction clustering method based on fuzzy theory
  • Image dimension reduction clustering method based on fuzzy theory

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0103] like figure 1 As shown, the fuzzy principal component dimensionality reduction clustering method includes the following steps:

[0104] Take the Control dataset as an example to introduce the method flow. The Control dataset has a total of 600 image samples with a dimension of 60, which are divided into 6 categories and reduced to the d' dimension. Then n=600, d=60, c=6, the sample matrix

[0105] ①Initialization which is

[0106] ②For the image data matrix XX T Perform eigendecomposition, and the eigenvectors corresponding to the first d′ largest eigenvalues ​​form a matrix U, initialization

[0107] ③Use the following formula to update the matrix V

[0108]

[0109] ④ Use the following formula to update the row vector m j , thus updating the matrix M

[0110]

[0111] ⑤Update matrix Y

[0112] for each row vector y i , which are calculated sequentially

[0113]

[0114] build function

[0115]

[0116] where (x) + =max(0,x).

[0117]...

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 an image dimension reduction clustering method based on a fuzzy theory, which comprises the following steps of: firstly initializing a projection matrix U, a membership matrix Y, a clustering center matrix M, a projected sample matrix V and a regularization parameter lambda, then alternately updating V, M and Y by adopting an alternate optimization algorithm, and repeatedly iterating until an objective function is converged to realize unsupervised data dimension reduction. According to the invention, the unsupervised method, namely the fuzzy principal component dimension reduction and clustering method (FPCPC), which can be used for carrying out dimension reduction and clustering at the same time, is realized. According to the method, the dimensionality reduction of the image data and the clustering in the subspace are simultaneously realized in one method, the efficiency is improved, and the loss of category information in the dimensionality reduction process of the image is reduced.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and in particular relates to an image dimension reduction clustering method. Background technique [0002] Dimensionality reduction and clustering are the two most popular algorithms in the field of machine learning. By projecting high-dimensional data into a low-dimensional space, dimensionality reduction eliminates redundant information and noise information in the original data, and retains the most important data features. It alleviates the problem of the "curse of dimensionality" brought about by high dimensions. In production practice, due to hardware failures, programming errors, program identification errors, etc., the obtained data often contains redundancy and noise. These redundancy and noise will not only cause the subsequent data processing operations to be complicated and slow, but also make the data processing more difficult. The result deviates from the true direction. ...

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 Applications(China)
IPC IPC(8): G06V10/762G06V10/77G06K9/62
CPCG06V10/763G06V10/77G06F18/23213
Inventor 王靖宇王林聂飞平李学龙
Owner NORTHWESTERN POLYTECHNICAL 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