摘要
提出了一种图像分割MPO算法,该算法将粒子群PSO算法与Otsu函数相结合,通过Otsu自适应函数来处理图像分割中的门槛值选择问题,使所求得的门槛值即为图像直方圆中组间变异数最大化的解﹒为增进计算速度,MPO算法将原PSO粒子群算法中的粒子群体最佳适应值移除,同时经过本算法改良,使新算法也能应用在多门槛值的选取问题﹒Matlab仿真实验表明:MPO算法是现有的门槛值算法Otsu与LEA运行速度的3.36~6.57倍,所以更具优势﹒
It is to propose an image segmentation MPO algorithm which combines the particle swarm optimization(PSO)algorithm and the Otsu function,the threshold selection problem in image segmentation is processed by the Otsu adaptive function,the threshold value is the solution of the maximum variation among the groups in the direct radius of the image.In order to improve the calculation speed the MPO algorithm removes the particle swarm optimization fitness value of the original PSO particle swarm optimization algorithm,and through the improved algorithm,the new algorithm can also be applied to the selection of multiple threshold values.The Matlab simulation results show that MPO algorithm is 3.36~6.57 times faster than Otsu and LEA,so it has more advantages.
作者
陈园
雷超阳
侯赞
刘军华
CHEN Yuan;LEI Chaoyang;HOU Zan;LIU Junhua(Department of Internet Engineering,Hunan Post and Telecommunication College,Changsha,Hunan 410015,China;Maintenance Company,State Grid Electric Power of Hunan,Changsha,Hunan 410004,China)
出处
《湖南城市学院学报（自然科学版）》
CAS
2018年第1期38-42,共5页
Journal of Hunan City University:Natural Science
基金
湖南省教育厅科研项目(16C720)
关键词
PSO
改良
MPO
图像分割
PSO
improvement
MPO
algorithm
image segmentation
