基于增强算术优化算法的多阈值图像分割  

Multi-Threshold Image Segmentation Based on Enhanced Arithmetic Optimization Algorithm

作　　者：吴荣生 WU Rongsheng(School of Electronic Information,Zhangzhou Institute of Technology,Zhangzhou 363000,China)

机构地区：[1]漳州职业技术学院电子信息学院,福建漳州363000

出　　处：《枣庄学院学报》2024年第5期47-54,共8页Journal of Zaozhuang University

基　　金：2024年漳州职业技术学院科研课题(zzykyk240015)。

摘　　要：针对传统多阈值图像分割方法中存在的分割质量一般、分割速度较慢等问题,提出一种基于改进的增强算术优化算法的多阈值图像分割方法。利用双重反向学习初始化种群,增强算法的搜索性能,将金枪鱼群优化算法的螺旋搜索策略引入到算术优化算法的加减策略中,帮助算法摆脱局部最优解。提出一种自适应余弦加速函数,更好地平衡算法的开发和探索能力。试验结果显示,提出的方法能够在提升算法收敛效率的同时分割出较好质量的图像。In view of the issues of poor segmentation quality and slow segmentation speed in traditional multi-threshold image segmentation methods,a multi-threshold image segmentation method based on enhanced arithmetic optimization algorithm is proposed.Double reverse learning is used to initialize the population,enhance the search performance of the algorithm.The spiral search strategy of tuna swarm optimization algorithm is introduced into the addition and subtraction strategy of arithmetic optimization algorithm to enhance the ability of jumping out of local optimization.An adaptive cosine acceleration function is proposed to better balance the development and exploration ability of the algorithm.The experimental results show that the method proposed is able to segment better quality images while improving the convergence efficiency of the algorithm.

关 键 词：算术优化算法 多阈值分割 反向学习 螺旋搜索 局部最优 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]