一种融合分水岭与ISODATA的岩心图像分割方法  

Core image segmentation method by combining watershed and ISODATA algorithm

作　　者：吴晓红[1] 王正勇[1] 罗代升[1] 

机构地区：[1]四川大学电子信息学院,成都610064

出　　处：《计算机应用》2008年第1期214-215,219,共3页journal of Computer Applications

基　　金：四川省科技攻关项目(05GG021-026-03)

摘　　要：分水岭算法用于图像分割,能获得封闭的和位置准确的轮廓,但容易造成过分割;ISODATA算法能将岩心砾石颗粒聚集成一类,但会使砾石目标边缘位置漂移。为克服以上两种方法的缺点,提出融合分水岭和ISODATA的图像分割方法。该方法首先利用分水岭算法得到过分割图像,ISODATA算法得到聚类图像,然后以ISODATA算法聚类的结果为依据,校正分水岭法的过分割问题。该方法用于砾岩图像的分割,取得了较好的实验效果。Watershed based image segmentation can make the edge close and precise, but has the drawback of over- segmentation. ISODATA can conglomerate gravel, but has the edge bias error. Combining watersheds and ISODATA, this paper proposed an image segmentation method to solve these problems. Firstly, watershed segmentation algorithm was applied to get the initial over-segmentation image, and Iterative Self-Organizing Data Analysis Technique Algorithm （ISODATA） was applied to get the conglomeration image. Then, labeled images and inverse wavelet transform were used together to get the full - resolution image. The results based on ISODATA algorithm were applied to solve over-segmentation problem caused by watershed. Experimental results show that it is a practicable method for the image segmentation.

关 键 词：分水岭 迭代自组织数据分析算法 过分割 砾岩图像 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]