融合正交学习和动态平衡的麻雀搜索算法及应用

打开文本图片集

关键词：麻雀搜索算法；正交学习；动态平衡；多阈值图像分割

中图分类号：TP301 文献标志码：A 文章编号：1001-3695（2025）08-020-2398-10

doi：10.19734/j. issn.1001-3695.2024.11.0526

Sparrow search algorithm integrating orthogonal learning and dynamic equilibrium and its application

He Hongyu¹，Pan Jiawen²，Qian Qian1† （1.Facultyofae&iongUesite&ngin;f Information& Electrical Enginering，China Agricultural University，Beijing1ooo83，China）

Abstract：Aiming atthe shortcomingsofthe sparrowsearch algorithm，suchasslowconvergencespeedandunbalancedsearch ability，this paper proposedasparrowsearch algorithm that integratedorthogonal learninganddynamicbalance（SSAOD）. Firstly，the SAODintroducedanorthogonalleaing strategytostrengthentheinformationtransmisionbetweenindividuals andimprove theconvergencespeedof thealgorithm.Then，thealgorithmadoptedadynamicbalancestrategytoenhance he globalexplorationabilityofthealgorithmintheearlyiterationandthelocalexploitationabilityinthelateriteration，soasto balancethetwosearchbehaviors.SSAODhadthebestoveralloptimizationability，betterstability，fasterconvergencespeed andhigheraccuracyonthe IEEECEC20 testfunctions.Ialsoappiedtheproposed algorithmtothemulti-threshold image segmentation problem.Asaresult，theimage details wereclearerandtheinformationwas more abundant，compared with the results of otheralgorithms.Theexperimentalresultsshowthatthe proposed SSAODnotonlyhasbeternumericaloptimization ability，but also has beter ability to solve practical problems.

KeyWords：sparrow search algorithm；orthogonal learning;dynamic balance;multi-threshold image segmentation

0 引言

全局优化问题广泛存在于科学研究、工程实践和经济金融中[1]，传统优化算法无法在有限的时间内快速解决离散，目标函数不可导，以及搜索空间复杂的优化问题[2.3]。（剩余15207字）