Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Supervised figure-based transductive data dimension-descending method

A data dimensionality reduction and supervising graph technology, applied in the field of image processing, can solve the problems of unsatisfactory classification and recognition effect, not considering the sample class label information, etc., to improve the data classification effect and improve the recognition performance.

Active Publication Date: 2016-06-15
XIDIAN UNIV
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method does not consider the class label information of the sample when projecting the k-nearest neighbor map
[0007] Therefore, the above-mentioned SELF and TCA methods are not ideal for classification and recognition of data after dimensionality reduction.

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
  • Supervised figure-based transductive data dimension-descending method
  • Supervised figure-based transductive data dimension-descending method
  • Supervised figure-based transductive data dimension-descending method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The specific implementation steps and effects of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0039] refer to figure 1 , the implementation steps of the present invention are as follows:

[0040] Step 1. Input the original image.

[0041] Input n=F×P original images, and after calibrating and aligning these images, they are cropped to the same size, where F is the number of original image categories, and P is the number of images of each category.

[0042] Step 2. Use the original image to obtain the original matrix X.

[0043] The gray feature value of each original image pixel is taken out by row, and arranged in sequence to form a d-dimensional row vector, forming an n×d matrix X', and normalizing each row of the matrix X', so that the matrix The sum of the elements of each row of X' is equal to 1: Among them, v' j is the jth row vector of matrix X', x' i is the row vector v' j i-th element, ...

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 transductive locality-preserving projection data dimension-descending method, which mainly solves the problem that the existing semi-supervised learning-based dimension-descending method is only applicable to the Euclidean distance-based figure composition and is not ideal in recognition result. The method comprises the steps of (1) inputting data and conducting the normalization treatment; (2) calculating the original matrix and the classification standard vector of normalized data; (3) calculating a first Laplacian matrix L based on the original data; (4) calculating a second Laplacian matrix L1 based on the classification standard vector; (5) calculating a similarity matrix S according to the first Laplacian matrix L and the second Laplacian matrix L1; (6) calculating an inter-class weight matrix Wc based on the classification standard vector; (7) constructing a generalized eigenvalue based on the similarity matrix S and the inter-class weight matrix Wc and solving to obtain a projection matrix E; (8) calculating a sample after the dimension-descending process based on the projection matrix E. According to the technical scheme of the invention, the feature extracting and the dimension descending of data are effectively realized. Meanwhile, the classification and recognition accuracy of data is improved. The method can be used for processing data and images.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a dimensionality reduction method for high-dimensional data, which can be used for data and computer image recognition. Background technique [0002] In recent years, with the development of computer technology and manufacturing industry, smart devices have become popular, such as smart phones, smart bracelets and so on. Almost every smart device has a large number of sensors to collect various data. The popularity of a large number of smart devices has been accompanied by an explosive growth of raw data. When we get more and more data, how to make full use of the information in the data has become the focus of academic research. Data dimensionality reduction is an effective means to solve these problems. Data dimensionality reduction aims to use popular computers to automatically explore the information in the original data and discover the essential features hidden ...

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): G06K9/00G06K9/62
CPCG06V40/172G06F18/24143
Inventor 王磊姬红兵王家俊朱明哲李丹萍
Owner XIDIAN 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
Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com