摘 要:图像分割是图像识别和计算机视觉至关重要的预处理。文章结合 PSO 算法搜索全局最优及 KMeans 算法对选取初始聚类中心的缺陷,使用 PSO 算法搜索并寻找全局最优,将搜寻值回代到 KMeans 聚类算法中对图像进行分割。结果表明,基 于 PSO-KMeans 快速图像分割算法模型具有更高的精确度和有效性。PSO 算法对 KMeans 算法的优化作用明显,有效对图片数据进行了分割,进一步优化了 KMeans 算法对图像分割的时间,提高了算法收敛的速度。
关键词:PSO 算法;KMeans 算法;PSO-KMeans 快速算法;收敛速度;图像分割
中图分类号:TP391.4 文献标识码:A 文章编号:2096-4706(2020)05-0079-04
Application of Fast Image Segmentation Algorithm Model Based on PSO-KMeans
CHEN Xingzhi,WANG Daiwen,LIU Naiyao,LE Wentao,HUANG Feixiang
(School of Science,Southwest University of Science and Technology,Mianyang 621010,China)
Abstract:Image segmentation is an important preprocessing of image recognition and computer vision. In this paper,PSO algorithm is used to search the global optimum and KMeans algorithm is used to select the initial cluster center. PSO algorithm is used to search and find the global optimum,and the search value is replaced by KMeans clustering algorithm to segment the image. The results show that the fast image segmentation algorithm model based on PSO-KMeans has higher accuracy and effectiveness. PSO algorithm plays an important role in the optimization of KMeans algorithm,effectively segmenting the image data,further optimizing the time of image segmentation of KMeans algorithm,and improving the convergence speed of the algorithm.
Keywords:PSO algorithm;KMeans algorithm;PSO-KMeans fast algorithm;convergence speed;image segmentation
基金项目:西南科技大学大学生创新基金项目(CX19-061)
参考消息:
[1] 孙越泓 . 基于粒子群优化算法的图像分割研究 [D]. 南京:南京理工大学,2010.
[2] 楚晓丽 .K-Means 聚类算法和人工鱼群算法应用于图像分割技术 [J]. 计算机系统应用,2013,22(4):92-94+103.
作者简介:
陈兴志(1997.11-),男,汉族,云南昆明人,本科,研究方向:应用数学;
王代文(1999.05-),男,汉族,四川绵阳人,本科,研究方向:应用数学;
刘乃瑶(1998.08-),女,汉族,吉林长春人,本科,研究方向:计算数学;
乐文涛(1998.12-),男,汉族,四川南充人,本科,研究方向:计算数学;
黄飞翔(1998.09-),男,彝族,云南昆明人,本科,研究方向:应用数学。