基于改进雁群算法的Otsu多阈值图像分割  

Otsu Multi-Threshold Image Segmentation Based on Improved Geese Swarm Algorithm

作　　者：郭业才[1,2] 赵涵优 GUO Ye-cai;ZHAO Han-you(School of Electronics&Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China;School of Electronics&Information Engineering,Wuxi University,Wuxi 214105,China)

机构地区：[1]南京信息工程大学电子与信息工程学院,江苏南京210044 [2]无锡学院电子信息工程学院,江苏无锡214105

出　　处：《中国电子科学研究院学报》2024年第7期622-633,646,共13页Journal of China Academy of Electronics and Information Technology

基　　金：国家自然科学基金资助项目(61673222)。

摘　　要：阈值分割是一种被广泛应用的图像分割技术。然而,传统的最大类间方差法(Otsu算法)在多阈值图像分割中面临着计算复杂度高、执行时间长及分割准确性不够等挑战。针对这些现象,提出一种基于改进雁群优化算法(CBLSGSO)的Otsu多阈值图像分割算法,该算法将Cubic混沌映射模型嵌入雁群算法初始化过程中,提高种群的多样性;提出多区域引导式结构,对种群动态切分并设计不同的进化机制,扩大种群寻优范围;引入自适应正余弦策略和蝴蝶算法搜索策略,提高算法的收敛精度,有效地平衡了算法的全局寻优能力和局部寻优能力。为验证改进后Otsu算法性能,选取ACC、Jaccard、Specificity、F1-score、FSIM、SSIM和PSNR等指标作为评价指标,并与近年来不同学者提出的图像分割算法进行实验对比,验证了算法的有效性。实验结果表明,基于改进雁群算法的Otsu图像分割法能更快速精确地解决复杂图像分割问题。Threshold segmentation is a widely used image segmentation technique.However,the traditional Otsu algorithm based on maximum interclass variance faces challenges in multi-threshold image segmentation such as high computational complexity,long execution time,and insufficient segmentation accuracy.To address these issues,a multi-threshold image segmentation algorithm based on improved geese swarm optimization algorithm(CBLSGSO)is proposed.This algorithm embeds the Cubic Chaos Mapping model into the initialization process of the gravitational search algorithm to enhance population diversity.A multi-region guided structure is introduced to dynamically divide the population and design different evolutionary mechanisms,expanding the population optimization range.Adaptive cosine strategy and butterfly algorithm search strategy are incorporated to improve the algorithm's convergence accuracy,effectively balancing global and local optimization capabilities.To verify the performance of the improved Otsu algorithm,ACC,Jaccard,Specificity,F1-score,FSIM,SSIM,and PSNR are selected as evaluation metrics and compared with image segmentation algorithms proposed by different scholars in recent years to validate the effectiveness of the algorithm.The experimental results demonstrate that the Otsu image segmentation method based on the improved algorithm can more quickly and accurately solve complex image segmentation problems.

关 键 词：多阈值图像分割 雁群算法 OTSU算法 混沌映射 蝴蝶优化算法 竞争机制 正余弦算法 

分 类 号：TN911.73[电子电信—通信与信息系统]