基于二维最小Tsallis交叉熵的图像阈值分割方法  

Image thresholding segmentation based on two-dimensional minimum Tsallis-cross entropy

作　　者：唐英干[1] 邸秋艳[1] 赵立兴[1] 关新平[1] 刘福才[1] 

机构地区：[1]燕山大学工业计算机控制工程河北省重点实验室,秦皇岛066004

出　　处：《物理学报》2009年第1期9-15,共7页Acta Physica Sinica

基　　金：国家杰出青年基金(批准号:60525303) ;燕山大学博士基金(批准号:B243);燕山大学科技发展基金(批准号:YDJJ200521)资助的课题~~

摘　　要：利用Tsallis熵的非广延性,提出了二维最小Tsallis交叉熵阈值分割方法.首先给出了二维Tsallis交叉熵的定义,并以最小二维Tsallis交叉熵为准则,利用粒子群优化算法来搜索最优二维阈值向量.该方法不仅进一步考虑了像素之间的空间邻域信息,而且考虑了目标和背景之间的相互关系,其分割性能优于基于Shannon熵的交叉熵阈值法和一维最小Tsallis交叉熵阈值法,并且具有很强的抗噪声能力.实验结果表明,该方法可以实现快速、准确的分割.Image thresholding segmentation method based on two-dimensional minimum Tsallis-cross entropy is proposed by utilizing the non-extensive property of Tsallis entropy in the paper.Firstly,the two-dimensional Tsallis-cross entropy is given,then the particle swarm optimization is used to search the best two-dimensional threshold vector by minimizing the two-dimensional Tsallis-cross entropy.The proposed method not only considers the spatial information of pixels,but also the interaction between the object and the background.Its segmentation performance is superior to thresholding methods using Shannon entropy and minimum one-dimensional Tsallis-cross entropy.Experimental results show that the proposed method can give good segmentation results with less computation time.

关 键 词：Tsallis交叉熵 二维直方图 粒子群优化算法 图像分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] P131[自动化与计算机技术—计算机科学与技术]