基于改进粒子群算法的二维阈值图像分割  

Two dimension threshold image segmentation based on improved particle swarm algorithm

作　　者：冯斌[1] 王璋[1] 孙俊[1] 

机构地区：[1]江南大学信息工程学院,江苏无锡214122

出　　处：《计算机应用研究》2008年第8期2402-2404,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(60474030)

摘　　要：二维Otsu方法同时考虑了图像的灰度信息和像素间的空间邻域信息,是一种有效的图像分割方法。针对二维Otsu方法计算量大的特点,采用量子粒子群算法来搜索最优二维阈值向量,每个粒子代表一个可行的二维阈值向量,通过各个粒子的飞行来获得最优阈值。结果表明,所提出的方法不仅能得到理想的分割结果,而且计算量大大减少,达到了快速分割的目的,便于二维Otsu方法的实时应用。2D Otsu method , which considers the gray information and spatial neighbor information between pixels in the image simultaneously , is an efficient image segmentation method. However , the computational burden of finding optimal thresh- old vector is very large for 2D Otsu method. Used a optimization method, such as quantum-behaved particle swarm optimization（QPSO） , to find the best 2D threshold vector , in which each particle represented a possible 2D threshold vector and the best 2D threshold was obtained through the flying among particles. Experimental results show that the proposed method can not only obtain ideal segmentation results but also decrease the computation cost reasonably , and it is suitable for real-time application.

关 键 词：图像分割 二维OTSU方法 粒子群算法 量子粒子群算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]