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

An Image Retrieval Method Based on Multi-Feature Fusion and Diffusion Process Reranking

A multi-feature fusion and diffusion process technology, applied in the field of image retrieval based on multi-feature fusion and diffusion process reordering, can solve the problem of low accuracy

Active Publication Date: 2020-08-14
YUNNAN NORMAL UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The invention provides an image retrieval method based on multi-feature fusion and diffusion process reordering, which is used to solve the problem of low accuracy of traditional image retrieval methods in CBIR, and realize efficient retrieval in large-scale natural image retrieval Target

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
  • An Image Retrieval Method Based on Multi-Feature Fusion and Diffusion Process Reranking
  • An Image Retrieval Method Based on Multi-Feature Fusion and Diffusion Process Reranking
  • An Image Retrieval Method Based on Multi-Feature Fusion and Diffusion Process Reranking

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] Embodiment 1: as figure 1 As shown, an image retrieval method based on multi-feature fusion and diffusion process reordering, this embodiment takes an image database composed of N (1000) images with a size of m×n (192×168) as an example, each The images are used as query images respectively, and the retrieval is completed by obtaining the similarity between each query image and other images in the database. The specific process includes: extracting the features of all images (Step1) and performing normalization and fusion (Step2), calculating the distance matrix between image features (you can get the similarity between each query image and other images in the database, the closer the distance The smaller the image, the more similar the image), and then introduce the diffusion process to optimize the distance matrix (Step3), and finally reorder it and complete the retrieval (Step4).

[0055] In the retrieval process, the present invention proposes an image feature extr...

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 multi-feature fusion and diffusion process reordering, comprising: Step1, image feature extraction; Step2, normalizing and fusing the image feature features extracted in step Step1; Step3, passing through steps The fusion feature of the image extracted in Step2 is optimized by the feature distance based on the diffusion process; Step4, reordering the optimized features in step Step3 and searching according to the rearrangement result. The method fusion feature proposed by the invention is easy to extract and has low complexity. The whole retrieval process does not require image segmentation and image classification training process, which can effectively solve the problem of low retrieval accuracy of the current traditional retrieval method based on the underlying visual features, and is better. It satisfies the actual needs of users for content-based image retrieval.

Description

technical field [0001] The invention relates to an image retrieval method based on multi-feature fusion and diffusion process reordering, and belongs to the related fields of computer vision, image processing, image understanding and the like. Background technique [0002] With the development of computer technology, more and more researchers pay attention to the related fields of computer vision. In recent years, image processing technology has been successfully applied in various industries, and content-based image retrieval (Content-based image retrieval, CBIR) is one of the main typical applications. CBIR stands for "Searching Images by Image". Different from the traditional search based on text keywords, CBIR focuses on the visual content of the image itself. The two key links of CBIR are image feature extraction and image similarity matching. [0003] Image features can be described from different visual perspectives such as color, texture, and shape. Based on this, ...

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 Patents(China)
IPC IPC(8): G06F16/583G06K9/46G06K9/62
CPCG06V10/44G06V10/56G06V10/467G06V10/464G06F18/23213G06F18/22
Inventor 周菊香甘健侯王俊
Owner YUNNAN NORMAL 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