基于区域的自动种子区域生长法的彩色图像分割算法  

Automatic SRG based region for color image segmentation

作　　者：孔俊[1] 王佳男[2] 谷文祥[1] 殷明浩[1] 

机构地区：[1]东北师范大学计算机学院,吉林长春130024 [2]东北师范大学网络信息中心,吉林长春130024

出　　处：《东北师大学报（自然科学版）》2008年第4期47-51,共5页Journal of Northeast Normal University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(60573067;60473042;60803102);东北师范大学青年自然科学基金资助项目(20070601)

摘　　要：提出了一种基于区域的自动种子区域生长法进行彩色图像分割的算法.该方法首先应用分水岭算法对图像进行初始化分割,形成过分割效果.然后从分水岭算法形成的区域出发,根据一定的规则自动选出一些区域作为种子区域,进行种子区域生长.与传统的种子区域生长法(SRG)以图像中的像素点作为种子进行生长的方法不同,本方法以区域作为种子并以区域作为生长单位进行区域生长.实验结果表明,该算法能够产生较好的分割效果及较快的分割速度.An automatic seeded region growing （SRG） based region algorithm for color image segmentation was provided in this paper. The method uses regions instead of pixels as the seeds of SRG. The architecture of the algorithm can be described as follows. First, the input RGB color image is transformed into HSI color space. Second, watershed segmentation was used to initialize the image. Third, the initial region seeds are automatically selected according two rules advanced by us. Fourth, the color image is segmented into regions. Finally, region-merging method is used to merge similar or small regions. Experimental results show that our algorithm can produce good results.

关 键 词：彩色图像分割 分水岭 自动种子区域生长 SRG 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]