首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种基于数学形态学的遥感影像分割方法
引用本文:纪松,张一鸣,龚辉.一种基于数学形态学的遥感影像分割方法[J].测绘工程,2007,16(4):33-36,40.
作者姓名:纪松  张一鸣  龚辉
作者单位:信息工程大学,测绘学院,河南,郑州,450052;信息工程大学,测绘学院,河南,郑州,450052;信息工程大学,测绘学院,河南,郑州,450052
基金项目:国家高技术研究发展计划(863计划)
摘    要:介绍形态学测地重建与开闭重建运算,提出基于微分形态学断面的影像分割方法。该方法首先生成原始影像的微分形态学断面影像,然后通过比较得到像点的最大微分断面值以及相对应的特征结构元素,最后以特征结构元素为层次,逐层分割影像。以IKONOS卫星影像为例,该方法能够有效地实现影像的分割,在分割较低反差的影像部分等方面具有优势。

关 键 词:测地重建  开闭重建  微分形态学断面  结构元素  影像分割
文章编号:1006-7949(2007)04-0033-04
收稿时间:2006-12-04
修稿时间:2006-12-04

An segmentation method of remotely sensed imagery based on mathematical morphology
JI Song,ZHANG Yi-ming,GONG Hui.An segmentation method of remotely sensed imagery based on mathematical morphology[J].Engineering of Surveying and Mapping,2007,16(4):33-36,40.
Authors:JI Song  ZHANG Yi-ming  GONG Hui
Institution:Institute of Surveying and Mapping, Information Engineering University, Zhengzhou 450052, China
Abstract:Morphological geodesic reconstruction and opening and closing reconstruction operations are introduced,and an image segmentation method based on differential morphological profile is proposed.This method first builds the differential morphological profile image of the original image,and then obtains the pixel's maximum differential profile value and its corresponding feature structure element,and finally,on the feature structure element layers,completes image segmentation process.Experimenting with IKONOS imagery,this method proves to be able to effectively segment images and have great advantages on segmentation images with low contrast areas.
Keywords:geodesic reconstruction  opening and closing reconstruction  differential morphological profile  structure element  image segmentation
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号