摘要
针对麻雀搜索算法(SSA)收敛速度慢,易陷入局部最优的问题,提出一种螺旋探索与自适应混合变异的麻雀搜索算法(SHSSA).首先,采用一种无限次折叠的ICMIC混沌初始化种群,增加种群多样性和遍历性,为全局寻优奠定基础;其次,融入一种螺旋探索策略,增强发现者探索未知区域的能力,提高算法的全局搜索性能;然后,提出一种基于精英差分和随机反向的混合变异策略,加快算法收敛速度,改善算法跳出局部最优的能力.基于12个基准测试函数的仿真结果表明,SHSSA与其余3种算法及2种改进的麻雀搜索算法相比,收敛速度更快、寻优精度更高,稳定性更强.最后,将SHSSA应用于多阈值图像分割中,实验结果表明,相较于基本SSA算法,SHSSA的分割速度和分割精度均得到了提升.
Aiming at the problem of slow convergence speed and easy to fall into local optimum of sparrow search algorithm(SSA),this paper proposed a sparrow search algorithm based on spiral search and adaptive hybrid mutation(SHSSA).Firstly,the algorithm used an infinite fold ICMIC chaos to initialize the population,which increases the diversity and ergodicity of the population and lays the foundation for global optimization;Secondly,the algorithm introduced a spiral search strategy to enhance the discoverer′s ability to explore unknown regions and improve the global search performance;Then,the algorithm introduced a hybrid mutation strategy based on elitist difference and random reverse to accelerate the convergence speed and improve the ability to jump out of local optimum.The simulation results based on 12 benchmark functions show that compared with the other three algorithms and two improved sparrow search algorithms,SHSSA has faster convergence speed,higher optimization accuracy and stronger stability.Finally,SHSSA is applied to multi threshold image segmentation.Experimental results show that compared with the basic SSA algorithm,the segmentation speed and accuracy of SHSSA are improved.
作者
陈功
曾国辉
黄勃
刘瑾
CHEN Gong;ZENG Guo-hui;HUANG Bo;LIU Jin(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2023年第4期779-786,共8页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61603242,61701296)资助。
关键词
麻雀搜索算法
混沌映射
螺旋探索
混合变异
sparrow search algorithm
chaotic map
spiral exploration
hybrid mutation
