基于形态分割的高分辨率遥感影像道路提取  

Road Extraction from High-resolution Remotely Sensed Image Based on Morphological Segmentation

作　　者：朱长青[1] 王耀革[2] 马秋禾[3] 史文中[4] 

机构地区：[1]中国科学院测量与地球物理研究所,湖北武汉430077 [2]解放军信息工程大学理学院,河南郑州450052 [3]解放军信息工程大学测绘学院,河南郑州450052 [4]香港理工大学土地测量与地理资讯学系

出　　处：《测绘学报》2004年第4期347-351,共5页Acta Geodaetica et Cartographica Sinica

基　　金：国家自然科学基金项目(40176032)

摘　　要：基于灰度形态学,提出一种从高分辨率遥感图像提取道路网络的方法。首先利用灰度形态特征对遥感影像进行分割,进而得到基本的道路网络轮廓。然后在此基础上,利用线段特征匹配方法提取道路网络。提出的方法能适应于从道路和背景区别不很清楚的遥感图像中提取道路。实验结果也表明,本文方法能有效地从遥感影像中提取道路网络。Based on grey level mathematical morphology, this paper presents a newly developed approach to extract road network from high-resolution remotely sensed image. First, the image is segmented based on grey level morphological characteristics, and basic road network can be obtained. Then final road network is extracted from the basic road network by line match method. The proposed approach in the paper can be adapted for road extraction from the remotely sensed image where road cannot be differentiated with background clearly. And the experiments also indicate that the proposed approach is efficient for extracting road network from remotely sensed image.

关 键 词：分割 遥感影像 高分辨率遥感图像 灰度 特征匹配 道路网络 轮廓 路提 基础 

分 类 号：P283.8[天文地球—地图制图学与地理信息工程] TP75[天文地球—测绘科学与技术]