一种结合空间信息的图像阈值分割新方法  

A NEW THRESHOLDING ALGORITHM USING SPACIAL INFORMATION

作　　者：夏勇[1] 赵荣椿[1] 

机构地区：[1]西北工业大学计算机科学与工程系,西安710072

出　　处：《中国体视学与图像分析》2002年第4期235-239,共5页Chinese Journal of Stereology and Image Analysis

摘　　要：针对基于灰度分布的图像阈值分割方法的不足,本文通过结合使用灰度级和空间相关信息将分类像素的特征空间从一维推广到二维,利用二维Otsu方法得到稳健的初始分割,然后使用松弛迭代的区域增长技术来判决边缘像素的归属。实验结果表明,本方法能够获得更好的阈值分割效果和更强的鲁棒性,并且大大降低了时间复杂度。Aimed at the shortage of the thresholding algorithms which based on the distribution of gray levers, this paper generalized the character space, in which the pixels are classified, from 1-D to 2-D by combining the distribution information of gray levers and the special information, then 2-D Otsu algorithm is used to obtain the original classification, finally a iterative region growing technique is applied to classify the edge pixels. The satisfied result of experiments proves that the presented algorithm has better performance, better robustness against noise and less time consuming than other methods in literature.

关 键 词：阈值化 OTSU方法 二维OTSU方法 区域增长技术 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]