Check patentability & draft patents in minutes with Patsnap Eureka AI!

A Color Image Retrieval Method Based on SIFT Seed Region Growth

A technology of region growth and color images, applied in the computer field, to achieve good robustness, improve efficiency and accuracy, and increase the effect of color invariance

Active Publication Date: 2016-11-30
SUZHOU VOCATIONAL UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The technical problem mainly solved by the present invention is to provide a color image retrieval method based on sift seed region growth, which can reduce the error problem caused by inaccurate image segmentation and make up for the deficiency of using a single feature to retrieve images

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
  • A Color Image Retrieval Method Based on SIFT Seed Region Growth
  • A Color Image Retrieval Method Based on SIFT Seed Region Growth
  • A Color Image Retrieval Method Based on SIFT Seed Region Growth

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Embodiment one: figure 1 It is a flow chart of a color image retrieval method based on SIFT seed region growth implemented by the present invention, and the image files are all color images.

[0045] 1. Select the sift seed point in the image to generate a sift seed point set E . Specific steps are as follows:

[0046] 1. Use the sift algorithm to detect local extremum points in the Gaussian scale space, then use the principal curvature of the Gaussian difference operator to delete edge response points in the local extremum points, and use the retained local extremum points as candidate feature points. Finally, the position of each candidate feature point in the original image is calculated.

[0047] 2. Use the Canny edge detection algorithm to detect the edge feature points of the image. Calculate the position of the pixel points within the 3×3 field of each edge feature point.

[0048] 3. If the position of the candidate feature point is not within the 3×3 area ...

Embodiment 2

[0080] Embodiment two: In order to illustrate the preference of using the region growing method based on the sift seed point to segment images in embodiment one, Figure 4 The method of the present invention is given. In the second step of the method of the present invention, the method of manually segmenting the image while other steps remain unchanged, and directly extracting the color sift feature and Hu invariant moment feature of the image without segmenting the image are used for image retrieval. The precision-recall trend chart of the method. The recall rate refers to the ratio of the number of correct images retrieved from the image library to the number of images in the image library that are consistent with the category of the image to be queried.

[0081] according to Figure 4 As a result, the method of the present invention uses the region growing method based on the sift seed point to segment the image, and then uses the KM algorithm to solve the optimal matchin...

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 color image retrieval method based on SIFT seed region growth, comprising the following steps: (1), for an image to be queried, selecting a SIFT seed point; (2), using a region growing method to divide the image into multiple regions (3), merge the regions according to the rules; (4), extract regional features for each region of the image; (5), calculate the similarity between the region of the image to be queried and the region of the image in the image library (6), construct the weight bipartite graph; Calculate the similarity between the image to be queried and the image in the image database; (7), arrange the images in the image database in descending order according to the similarity with the image to be queried. Through the above method, the present invention can reduce the error problem caused by inaccurate image segmentation and make up for the deficiency of using a single feature to retrieve images.

Description

technical field [0001] The invention relates to the computer field, in particular to a color image retrieval method based on SIFT seed region growth. Background technique [0002] As a kind of multimedia information with rich connotation and intuitive expression, image is favored by people. More and more business activities, business transactions and information representations contain image data. Especially in the popular electronic shopping on the Internet, the information of products is basically shown to users in the form of images. [0003] Traditional image retrieval methods use the underlying features of images, such as color, texture, shape, and spatial relationship, and these features have their own shortcomings. The color feature is not sensitive to changes in the direction and size of the image or image area, and cannot capture the local features of the object in the image well, nor can it express the information of the color space distribution; The texture ref...

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): G06F17/30G06T7/00
Inventor 杨元峰李金祥鲜学丰廖黎莉李亚琴
Owner SUZHOU VOCATIONAL UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com