Unlock instant, AI-driven research and patent intelligence for your innovation.

Improved Multi-threshold Image Segmentation Method Optimized by Reverse Harmony Search

An image segmentation and multi-threshold technology, applied in image analysis, image data processing, instruments, etc., can solve the problem of low segmentation accuracy and achieve the effect of improving accuracy

Active Publication Date: 2019-05-14
JIANGXI UNIV OF SCI & TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional harmony search algorithm is prone to fall into local optimum when optimizing multi-threshold image segmentation, and the segmentation accuracy is not high.

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
  • Improved Multi-threshold Image Segmentation Method Optimized by Reverse Harmony Search
  • Improved Multi-threshold Image Segmentation Method Optimized by Reverse Harmony Search
  • Improved Multi-threshold Image Segmentation Method Optimized by Reverse Harmony Search

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0046] Step 1, input the image HM to be segmented such as figure 1 shown;

[0047] Step 2, the user initializes parameters, the initialization parameters include the number of segmentation thresholds D=3, the size of the harmony library Popsize=50, the mutation rate Pmu=0.15, and the maximum number of evaluations MAX_FEs=1000;

[0048] Step 3, let the current evolution algebra t=0, the current evaluation times FEs=0;

[0049] Step 4, set the lower bound LB of the D segmentation thresholds j and upper bound UB j , where the dimension subscripts j=1,2,...,D;

[0050] Step 5, Randomly generate the initial harmony library where individual subscripts i=1,2,...,Popsize, and for the harmony library P t The i-th individual in , stores D segmentation thresholds;

[0051] Step 6, calculate the harmony library P t The fitness value of each individual in ;

[0052] Step 7, let the current evaluation times FEs=FEs+Popsize;

[0053] Step 8, save the harmony library P t The best...

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-threshold image segmentation method using improved reverse harmony search for optimization. The method adopts an improved reverse harmony search algorithm to optimize thresholds of image segmentation. In the improved reverse harmony search algorithm, first a harmony search strategy is executed to generate a test individual, a combined reverse learning strategy is executed to generate a reverse individual of the test individual, then the wining individual is selected from the test individual and the reverse individual, and the selected winning individual competes with the worst individual in a harmony library. Through execution of the combined reverse learning strategy, the multi-threshold image segmentation method provided by the invention can improve the multi-threshold image segmentation precision.

Description

technical field [0001] The invention relates to the field of image segmentation, in particular to a multi-threshold image segmentation method for improving reverse harmony search optimization. Background technique [0002] Multi-threshold image segmentation is a common image segmentation method, which has been successfully applied in medical imaging, face recognition, iris recognition, optical character recognition, fingerprint recognition and so on. The idea of ​​multi-threshold image segmentation method is to find several thresholds according to a given digital image to divide the pixels of the image into several categories so that the objective function can reach the optimal value. Multi-threshold image segmentation is essentially an optimization problem. The core of optimizing multi-threshold image segmentation is how to efficiently find the appropriate threshold. However, the traditional search methods use brute force to search for the optimal segmentation threshold, ...

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): G06T7/11G06T7/136
Inventor 郭肇禄杨火根章银娥王洋尹宝勇鄢化彪余法红
Owner JIANGXI UNIV OF SCI & TECH