Affiliations:
Xidian University, School of Mathematics and Statistics, Xi’an, China
Correspondence:
[*]
Corresponding author. Shengtao Jiang, Xidian University, School of Mathematics and Statistics, 266 Xinglong Section of Xifeng Road, Xi’an 710126, China. Tel.: +86 18392396160; Email: stjiang@stu.xidian.edu.cn.
Abstract: A generalized fuzzy entropy based on double adaptive ant colony algorithm for image thresholding segmentation is proposed. The new algorithm first attempts to propose the adaptive pheromone concentration at the initial time and the adaptive global updating rules, which uses the double adaptive mechanism to automatically select the generalized fuzzy entropy parameters. The threshold of the image is obtained by introducing the parameters into the complement of the generalized fuzzy entropy, and then the optimal segmentation of the image is obtained. Compared with the existing image thresholding segmentation algorithms, in most cases, simulating results indicate that the new algorithm has less background information and clearer target information. In addition, it is superior to the existing algorithms in performance and greatly improves the stability and convergence speed.
