Unseen image feature migration method based on self-organizing graph constraint non-negative matrix factorization

A non-negative matrix factorization and image feature technology, applied in the field of image processing, can solve problems such as limiting feature migration performance, and achieve the effect of reducing possibility, speeding up the process of iteration, and improving robustness

Active Publication Date: 2021-01-15
CHINA JILIANG UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the graph structure used in this method comes from the original image space that is sensitive to noise, not a more robust feature space, which limits the performance of feature transfer.

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
  • Unseen image feature migration method based on self-organizing graph constraint non-negative matrix factorization
  • Unseen image feature migration method based on self-organizing graph constraint non-negative matrix factorization
  • Unseen image feature migration method based on self-organizing graph constraint non-negative matrix factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] refer to figure 1 , the specific implementation steps of the present invention are as follows:

[0018] Step 1, separately from the auxiliary field D s and target domain D t Select several image samples to form a training sample set Among them, d represents the dimensionality of the image sample, n s and n t represent the number of training samples selected from the auxiliary domain and the target domain, respectively.

[0019] Step 2, according to the image sample Y of the auxiliary domain s and label information L s , initialize the base matrix A and feature matrix S of the auxiliary domain and the target domain, where, c is the number of sample categories in the auxiliary domain, rank r=c.

[0020] 2a) The base matrix A is initialized with a random non-negative number between 0-1;

[0021] 2b) The first to n_s rows of the feature matrix S are initialized with L_s, and the n_s to n_s+n_t rows are initialized with the following formula:

[0022]

[0023]...

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 feature migration method based on self-organizing graph constraint non-negative matrix factorization. The problem that image non-negative feature migration is not seenacross fields is mainly solved. The method comprises the following steps: (1) respectively selecting a plurality of image samples from an auxiliary domain Ds and a target domain Dt to form a trainingsample set; (2) initializing a basis matrix A and a feature matrix S of an auxiliary field and a target field; (3) calculating a basis matrix graph GA and a feature matrix graph GS; (4) setting the number of iterations T, and optimizing by using an iterative strategy to obtain a final basis matrix A and a final feature matrix S; and (5) calculating the label of the test image through the basis matrix A. Characteristic self-organizing graph constraints are adopted, characteristic regression parameters are introduced, the characteristic migration robustness of an unseen image is enhanced, and the method can be used for solving the problem of image non-negative characteristic migration.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to an image feature transfer learning method, and can be used for image non-negative feature extraction and analysis. Background technique [0002] Non-negative matrix factorization is a common image feature extraction method, which decomposes the original image set matrix into two low-rank non-negative matrices to obtain the base matrix and feature matrix of the image set. Since the non-negative matrix factorization can describe the local non-negative features of the image, it is widely used in various data feature extraction and analysis tasks. With the development of data scale, data in different fields often share some information, and extracting cross-field shared information to improve data feature extraction in the target field is the research focus of non-negative matrix factorization in cross-field feature transfer. Therefore, feature transfer based on non-negative matr...

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): G06T3/00G06K9/62
CPCG06F18/214G06F18/241G06T3/04
Inventor 朱文杰
Owner CHINA JILIANG 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