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

Image Retrieval Method Based on Sparse Nonnegative Matrix Factorization

A non-negative matrix decomposition and image retrieval technology, applied in the field of image retrieval based on sparse non-negative matrix decomposition, which can solve the problem of ignoring sparsity control and so on

Inactive Publication Date: 2011-12-07
ZHEJIANG UNIV
View PDF0 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, although the method proposed in [1] considers the control of sparsity in matrix decomposition, this method is limited to the use of a single data source (that is, a single matrix); while the method proposed in [2] uses multiple data sources knowledge to assist image retrieval, but ignores the control of sparsity

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 Retrieval Method Based on Sparse Nonnegative Matrix Factorization
  • Image Retrieval Method Based on Sparse Nonnegative Matrix Factorization
  • Image Retrieval Method Based on Sparse Nonnegative Matrix Factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0078] figure 2 The comparison results between the image retrieval of the present invention and other methods are given.

[0079] 1) Querying and extracting the image and its accompanying text of the retrieval result respectively under two different image data sources according to two different keyword sets to form the first image data set D 1 and a second image dataset D 2 ;

[0080] 2) Extract the tags in the accompanying text, and filter to form a vocabulary according to word frequency;

[0081] 3) For each image data set, use the association relationship between the label and the image to form an association matrix X between the label and the image 1 、X 2 ;

[0082] 4) Use the sparse non-negative matrix factorization to analyze the correlation matrix obtained in step 3) to obtain the shared subspace w of data from different sources 12 and the corresponding independent subspace w 1 、w 2 ;

[0083] 5) Carry out a series of queries on the images on a certain data so...

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 retrieval method based on sparse non-negative matrix decomposition. It includes the following steps: 1) Query and extract the image and accompanying text of the retrieval results under two different image data sources; 2) Extract the tags in the accompanying text, and filter the results according to word frequency to form a vocabulary; 3) For each Image set, using the association relationship between labels and images to form an association matrix between labels and images; 4) Using sparse non-negative matrix decomposition to analyze the association matrix obtained in step 3), to obtain the shared subspaces of different sources of data and their corresponding independent subspaces Subspace; 5) The user sends a retrieval request for images on a data source, forms a query vector and maps it to the corresponding subspace of the data source, calculates the similarity with all images and sorts them, and returns the top N most similar images. The present invention makes full use of the associated knowledge of tags and images under multiple data sources, performs migration learning through sparse non-negative matrix decomposition, and improves the accuracy of image retrieval on target data sources.

Description

technical field [0001] The invention relates to the field of image retrieval, in particular to an image retrieval method based on sparse non-negative matrix decomposition. Background technique [0002] As one of the characteristics of web 2.0, social tags are becoming more and more popular. On websites such as Flickr, YouTube, and Del.icio.us, users can mark photos, videos, web pages, etc., and retrieve related resources according to their interests. However, there are problems such as noise, ambiguity, and subjectivity in the tags added by users, and it is impossible to obtain satisfactory results by directly using the tags marked by users to retrieve resources. Therefore, how to improve the effect of image retrieval based on existing labels is a hot issue in current research. In recent years, many methods have been proposed for this problem. However, these methods have a common limitation, that is, most of them use information from a single data source and ignore the ef...

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): G06F17/30
Inventor 吴飞马帅邵健肖俊
Owner ZHEJIANG 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