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

Multi-target image segmentation C-V method based on area division and gradient guiding

A C-V, multi-target technology, applied in image analysis, image enhancement, image data processing, etc., can solve the problems of target over-segmentation, missing targets, slow evolution, etc., to overcome sensitivity, avoid omission, improve accuracy and efficiency effect

Inactive Publication Date: 2016-06-22
LIAONING NORMAL UNIVERSITY
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it is difficult to effectively deal with objects with uneven gray levels, especially for multi-target segmentation, it is easy to miss the target or over-segment the target
In addition, this method is too rough to deal with the rich information of image texture, and the evolution speed is slow.

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
  • Multi-target image segmentation C-V method based on area division and gradient guiding
  • Multi-target image segmentation C-V method based on area division and gradient guiding
  • Multi-target image segmentation C-V method based on area division and gradient guiding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] A C-V method for image multi-target segmentation based on region division and gradient guidance, comprising the following steps:

[0022] Step 1. Quickly determine the approximate outline of the target boundary by the maximum between-class variance method;

[0023] Step 2. Taking the rough contour of the determined target boundary as the initial contour, the evolution process of converting the image target segmentation into a level set:

[0024] Step 2.1 performs binarization processing on the image based on the maximum inter-class variance method, and roughly distinguishes the target area from the background area, wherein the "0" area is the approximate background area, and the "1" area is the approximate target area;

[0025] Step 2.2 For the binarized image, determine the approximate outline of the target area by calculating the difference between each pixel and its adjacent pixel gray values ​​in the four directions above, below, left, and right, that is, if the fou...

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 multi-target image segmentation level set C-V method based on area division and gradient guiding. In the method, through a maximum between-cluster variance method, a general outline of a target boundary is rapidly determined, and then an outline line is used as an initial outline, an evolution process of using a multi-target segmentation level set C-V model based on the gradient guiding to carry out multi-target segmentation on an image so as to convert into a single level set is included, and then through the self-adapting gradient guiding, an evolution process of an outline line level set is controlled to realize segmentation to multiple targets. On a basis of considering global image information, through the general outline of each target, a local characteristic of the target is considered so that an omission phenomenon of the targets is effectively avoided. And through the gradient guiding, target segmentation precision and efficiency are increased. Besides, in the invention, a sensitivity of a traditional model to an initial evolution curve is overcome.

Description

technical field [0001] The invention belongs to a digital image processing method, in particular to a C-V method for image multi-object segmentation based on region division and gradient guidance. Background technique [0002] Image segmentation is to separate the "target" area in the image from other "background" areas, and it has been paid attention to as a basic work of image processing and analysis. However, due to the "morbid" characteristics of the image itself and the different information characteristics contained in different images, the segmentation effects brought by different segmentation methods are different. So far, it is difficult to find a suitable method for various images. efficient method for segmentation. In recent years, the image segmentation method based on the active contour model (Active ContourModel, ACM) has made good research progress. The basic idea is to control an initialized closed contour curve to move inward or outward through the energy f...

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): G06T7/00
CPCG06T2207/20116
Inventor 王相海王金玲万宇赵婉彤
Owner LIAONING NORMAL UNIVERSITY
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

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