一种改进的粒子群爬山优化图像分割方法  

An Improved Particle Swarm Optimization Hill-climbing Image Segmentation Method

作　　者：孙光灵 吴倩[1] 卫星 SUN Guangling;WU Qian;WEI Xing(School of Electronic and Information Engineering,Anhui Jianzhu University,Hefei 230601,China;Intelligent Interconnected Systems Laboratory of Anhui Province,Hefei 230009,China;School of Computer Science and Information Engineering,Hefei University of Technology,Hefei 230601,China)

机构地区：[1]安徽建筑大学电子与信息工程学院,安徽合肥230601 [2]合肥工业大学“智能互联系统安徽省实验室”,安徽合肥230009 [3]合肥工业大学计算机与信息学院,安徽合肥230601

出　　处：《佳木斯大学学报（自然科学版）》2023年第1期11-15,共5页Journal of Jiamusi University：Natural Science Edition

基　　金：国家自然科学基金资助项目(62001004);安徽省高校协同创新项目(GXXT-2021-024);合肥工业大学开放基金(PA2021AKSK0107)。

摘　　要：针对粒子群优化算法在图像分割中存在算法搜索能力不足并且分割不够精准的问题,文章提出了一种改进的粒子群爬山优化图像分割方法。文章首先引入爬山算法,检测粒子群搜索空间中多个全局极值点,增强粒子群算法的局部搜索能力;其次将算法作用于图像中生成K个峰值的三维直方图,根据欧几里德距离,将每个像素分配给最近的峰值从而分割图像;利用标准图像数据集进行实验,并与其他基于粒子群优化的图像分割方法对比分析,该算法在图像分割的视觉效果和5种常用的客观评价指标都具有更优越的性能。In order to solve the problem of insufficient searching ability and segmentation accuracy of particle swarm optimization algorithm in image segmentation, this paper proposes an image segmentation method based on particle swarm optimization for mountain climbing. Firstly, mountain climbing algorithm is introduced to detect multiple global extreme points in the particle swarm search space to enhance the local search ability of particle swarm algorithm. Secondly, the algorithm is applied to the image to generate a three-dimensional histogram of K peak values. According to Euclidean distance, each pixel is assigned to the nearest peak to segment the image. Compared with other image segmentation methods based on particle swarm optimization(PSO), this algorithm has better performance in the visual effect of image segmentation and five common objective evaluation indexes.

关 键 词：粒子群 爬山算法 局部搜索 三维直方图 全局极值 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]