A fast SAR image segmentation method based on improved chicken swarm optimization algorithm |
| |
Authors: | Jianhui Liang Lifang Wang Miao Ma Jian Zhang |
| |
Affiliation: | 1.School of Computer Science,Northwestern Polytechnical University,Xi’ an,China;2.Institute of Tropical Agriculture and Forestry,HaiNan University,Dan zhou,China;3.School of Computer Science,Shaanxi Normal University,Xi’ an,China |
| |
Abstract: | Severe speckle noise existed in synthetic aperture radar (SAR) image presents a challenge to image segmentation. Though some traditional segmentation methods for SAR image have some success, most of them fail to consider segmentation effects and segmentation speed at the same time. In this paper, we propose a novel method of SAR image fast segmentation which is based on an improved chicken swarm optimization algorithm. In this method, the positions of the whole chicken swarm are firstly initialized in a narrowed foraging space. Secondly, the grey entropy model is selected as the fitness function of the improved chicken swarm optimization algorithm. Hence, the optimal threshold value is located gradually and quickly by virtue of the foraging behaviors of chicken swarm with a hierarchal order. Experimental results show that our method is superior to some segmentation methods based on genetic algorithm, artificial fish swarm algorithm in convergence, stability and segmentation effects. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|