一种基于淹没的分水岭变换算法  

Watershed Transform Based on Flooding

作　　者：蒋志勇[1] 陈晓玲[1,2] 秦前清[1] 

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,武汉市珞430079 [2]江西师范大学鄱阳湖湿地与流域研究教育部重点实验室,南昌市330022

出　　处：《武汉大学学报（信息科学版）》2009年第5期509-513,共5页Geomatics and Information Science of Wuhan University

基　　金：国家973计划资助项目(2006CB701300);国家863计划资助项目(2007AA12Z161);国家创新研究群体科学基金资助项目(40721001);国家“十一五”计划资助项目(2007BAC23B05);国家科技部中德国际合作计划资助项目(2006DFB91920);国家自然科学基金资助项目(40676094,40523005,40601055);上海市重点学科开放研究基金资助项目(T0102);2006/2007年度中爱科技合作研究计划资助项目

摘　　要：针对分水岭变换产生的过分割问题以及标记点选取可能导致的欠分割问题,提出了一种基于变换序列滤波和梯度淹没的分水岭变换算法。利用基于重构的变换序列滤波来平滑影像,消除区域内部差异;利用面积淹没和深度淹没处理过渡区域在梯度影像上造成的极小区域,然后对其进行分水岭变换。该算法可以有效地控制分水岭变换的过程,解决分水岭变换的过分割和欠分割问题。Watershed transform is the main tool of mathematical morphology used for image segmentation. This method applied directly to gradient image causes severe over-segmentation. We present an improved watershed transform based on morphological reconstruction and flooding by the analysis of the reasons of over segmentation produced by watershed and the possible under-segmentation due to watershed-plus-marker. The method proposed here is divided into three steps. （1） the alternate sequential filters based on reconstruction are used to reduce some difference inside objects;（2） the area oriented flooding and depth oriented flood ing are applied to gradient image in order to fill some local minima that does not correspond to maximum and minima in original image; （3） the processed gradient image is segmented by the watershed transform. The method proposed here can effectively control watershed transform and reduce over-segmentation and under-segmentation.

关 键 词：重构 变换序列滤波 淹没 分水岭变换 

分 类 号：P237.4[天文地球—摄影测量与遥感] TP753[天文地球—测绘科学与技术]