基于Renyi熵与PSO算法的图像多级阈值分割  

Image multi-thresholding using Renyi entropy and PSO

作　　者：聂方彦[1] 张平凤[1] 潘梅森[1] 张奋[1] 

机构地区：[1]湖南文理学院图形图像处理技术研究所,湖南常德415000

出　　处：《湖南文理学院学报（自然科学版）》2013年第3期44-48,52,共6页Journal of Hunan University of Arts and Science(Science and Technology)

基　　金：湖南省普通高校青年骨干教师培养对象资助项目(湘教通(2011)388号);湖南文理学院博士科研启动基金资助项目(107|13101022)

摘　　要：在图像阈值分割方法中,Renyi熵法因其显著效能而得到大量应用.为了更好地发挥Renyi熵在图像分割中的应用,提出把Renyi熵法扩展到图像多级阈值化问题.然而,由于计算时间复杂度上的高要求,很难把这种有效的技术推广到复杂图像多级阈值化问题.为减少本方法的计算时间,应用粒子群优化算法实施最佳阈值的搜索.实验结果表明,本方法能有效地对图像进行多级分割,并且显著降低计算时间.In image segmentation, Renyi-based thresholding method has obtained widely application because of its remarkable effectiveness. In order to develop the latent ability of Renyi-based method in image segmentation, the method was extended to multi-thresholding field. However, due to the time complexity, the Renyi entropy-based method was very difficult extended to multi-thresholding scenario straightly. To overcome this problem, a fast multi-thresholding method combined with the particle swarm optimization algorithm for complex image segmentation was proposed based on Renyi entropy. The experimental results show that the proposed method can reduce the computation time greatly, and obtain ideal segmentation result.

关 键 词：图像分割 多级阈值化 RENYI熵 粒子群优化 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]