改进粒子群算法的图像分割阈值方法研究  

image segmentation threshold method of improved particle swarm algorithm

作　　者：刘笃晋[1] 

机构地区：[1]四川文理学院计算机科学系,四川达州635000

出　　处：《中国科技信息》2013年第15期54-54,58,共2页China Science and Technology Information

基　　金：四川文理学院研究基金(2012Z001Y)

摘　　要：图像分割在图像处理领域里的作用至关重要,当前图像分割方法的一类重要方法是源于粒子群算法的图像分割方法,本文对源于改进粒子群算法的三类图像分割方法,即单阈值图像分割方法、二维阈值图像分割方法和多阈值图像分割方法进行了全面研究,通过研究指出单阈值图像分割方法效果相对来说是较差的,二维阈值图像分割方法比单阈值图像分割方法好,多阈值图像分割方法效果最好,并对每一类方法的不足以及以后要解决的问题也作了详细分析,同时也通过研究指出了图像分割技术未来的发展方向将是多种方法相结合才能得到较好的效果。Image segmentation is crucial in the field of image processing, in the current image segmentation methods there is an important class that is derived from the particle swarm algorithm for image segmentation method, this paper comes from three types of improved PSO algorithm image segmentation method, this paper conduct a comprehensive study for a single threshold image segmentation method, two-dimensional threshold segmentation method and multi--threshold image segmentation method , the research indicates single-threshold image segmentation method effect is relatively poor, two dimensional threshold image segmentation method than the single threshold image segmentation method is good, multi--threshold image segmentation method works best, and also made a detailed analysis about each class method deficiencies and future direction of development as well as future problems to be solved, but also pointed out the future direction of the image segmentation technology by studying,that will be combined with a variety of methods to get better results.

关 键 词：图像分割 异步粒子群算法 量子粒子群算法 小生境粒子群算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]