多策略改进的沙猫群优化低照度图像增强算法OA
Multi-Strategy Improved Sand Cat Swarm Optimization Low-Light Image Enhancement Algorithm
针对沙猫群优化算法在种群初始阶段多样性不足、收敛速度慢以及收敛效率低等问题,文章提出了一种基于混合多策略的改进沙猫群优化算法.首先,引入佳点集初始化策略来改进初始种群的生成机制,从而丰富种群多样性.其次,加入非线性收敛因子以平衡全局搜索和局部开发,提高算法的整体收敛效率.然后,结合周期性变异策略来提升算法在搜索阶段跳出局部最优的能力,并采用动态螺旋搜索策略以进一步加快收敛速度.最后,选取16个基准测试函数来开展对比实验,并通过Wilcoxon秩和检验完成分析.实验结果表明,在低照度图像增强场景中,所提算法的信息熵、标准差和平均梯度等参数相较于对比算法平均提高了0.075 6、1.018 6和0.230 2,充分验证了所提算法具备更强的寻优能力,能够有效优化图像处理效果,并输出画质更优的视觉图像.
Aiming at the problems of insufficient diversity,slow convergence speed and low convergence efficiency in the initial population stage of the sand cat swarm optimization algorithm,an improved sand cat swarm optimization algorithm based on hybrid multi-strategies is proposed in this paper.Firstly,the good point set initialization strategy is introduced to im-prove the generation mechanism of the initial population and enrich the diversity of the population.Secondly,a nonlinear con-vergence factor is incorporated to balance global search and local exploitation,which improves the overall convergence effi-ciency of the algorithm.The periodic variation strategy is utilized to enhance the ability of the algorithm to escape from local optimal traps in the searching stage.And the dynamic spiral search strategy is used to further accelerate the convergence speed.Finally,16 benchmark test functions are selected for comparative experiments,and the results are analyzed using the Wilcoxon rank-sum test.Experimental results demonstrate that in low-light image enhancement scenarios,the proposed algo-rithm outperforms the comparison algorithms by an average of 0.075 6 in information entropy,1.018 6 in standard deviation,and 0.230 2 in average gradient,respectively.This fully verifies that the proposed algorithm has stronger optimization capabili-ty,can effectively improve image processing,and outputs visual images with superior quality.
张柳莹;纪娟娟;戚茗方;庞子祥;杜明;查申龙
安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133安庆师范大学 电子信息与集成电路学院,安徽 安庆 246133
信息技术与安全科学
智能优化算法沙猫群优化算法非完全beta函数佳点集初始化低照度图像增强
intelligent optimization algorithmsand cat swarm optimization algorithmincomplete beta functiongood point set initializationlow illumination image enhancement
《安庆师范大学学报(自然科学版)》 2026 (2)
47-56,10
国家自然科学基金项目(62205005)和安徽省自然科学基金青年基金项目(2108085QD182)
评论