摘要
提出了一种多目标混合优化的阈值图像分割算法。该方法以类间方差函数和模糊熵函数为待优化目标函数,为了改善粒子群算法在迭代后期陷入局部最优的问题,在粒子群算法中引入多元宇宙优化算法并产生一组非支配解集;采用混沌搜索策略进行搜索,以更有效地逼近最优阈值;通过类间差异和类内差异的加权比值来选取最优解。仿真结果表明,相较于Otsu算法、多目标粒子群算法以及多元宇宙优化算法,算法的分割准确率较高。
A threshold image segmentation algorithm based on multi-objective hybrid optimization is proposed.The inter-class variance function and the fuzzy entropy function are taken as the objective function.In order to solve the problem that the particle swarm optimization algorithm falls into local optimum in the late iteration,a multi-verse optimization algorithm is introduced into the particle swarm optimization algorithm to generate a set of non-dominated solution sets.In addition,in order to enhance the accuracy of the algorithm,chaotic search strategy is adopted around the solution to help the algorithm approach the optimal threshold more effectively.Finally,the optimal solution is selected by weighted ratio of inter-class and intra-class differences.The simulation results show that compared with Otsu algorithm,multi-objective particle swarm optimization algorithm and multi-objective multi-verse optimization algorithm,the proposed algorithm can achieve better segmentation accuracy.
作者
赵凤
孔令润
ZHAO Feng;KONG Lingrun(School of Communication and Information Engineering,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;Key Laboratory of Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xi'an 710121,China)
出处
《西安邮电大学学报》
2019年第5期15-25,共11页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金资助项目(61571361,61102095,61671377)
西安邮电大学“西邮新星”团队支持计划资助项目(xyt2016-01)
关键词
阈值分割
粒子群优化
多元宇宙优化
混沌搜索
混合优化
thresholding segmentation
particle swarm optimization algorithm
multi-verse optimization
chaotic search
hybrid optimization
