摘要遥感图像是地理数据系统重要的数据源之一。随着遥感技术的发展，影像的分辨率在不断提高，从高分辨率遥感图像中提取道路已成为当前的研究热点之一。本文对高分辨率遥感图像道路提取方法进行了深入的研究。文中主要是以数学形态学方法为基础，并且结合了路径形态学方法，对现有的道路提取方法做出来了一些改进。
形态建筑指数（MBI）的提出最初用来检测建筑物。MBI是根据建筑物的形状，方向和大小来确定建筑物。本实验对MBI进行了改进，将道路的一些特征，如道路的宽度和方向，应用到MBI算法中。在MBI算法的过程中，运用数学形态学的知识得到初步的道路图像，但是该图像道路信息不明显，所以使用路径形态学方法根据道路的几何信息突出道路。通过对实验效果的分析，证实了此改进算法的实用性。30159
毕业论文关键词 高分辨率遥感影像，道路提取，数学形态学，形态建筑指数
外文摘要
Title Extracting road from high resolution remote sensing image based on Mathematical Morphology
Abstract
Remote sensing image is one of the important data source for the geographic data system. With the development of remote sensing technology, the resolution of the image is improving and extracting roads from high-resolution remote sensing image has become one of the hotspots. In this paper, our research mainly focuses on the high resolution remote sensing image road extraction method. The road extraction process is implemented based on mathematical morphology combined with path morphology approach.
Morphological building index (MBI) was presented by researchers to detect buildings in high resolution remote sensing images, which is based on the shape, face and size of the building. In this dissertation, we use the characteristics of the road, such as the width and direction of the road, to improve the MBI algorithm. In the procedure of MBI algorithm, mathematical morphology is used to get preliminary road regions. As the road information is not obvious, we use path morphology that based on geometry information to highlight the candidate road areas. Through analyzing the result of the experiment, the practicability of the improved algorithm has been validated.
Keyword: high-resolution image, road extraction, mathematical morphology, morphological building index
目 次
1 绪论 1
1.1 选题背景及意义 1
1.2 国内外高分辨率遥感影像道路提取研究现状 1
1.2.1 半自动道路提取 2
1.2.2 自动道路提取 3
1.3 论文的主要研究内容和组织结构 4
2 数学形态学的基础理论 5
2.1 数学形态学基本介绍 5
2.2 二值形态学理论 5
2.2.1 腐蚀与膨胀运算 6
2.2.2 开、闭运算 7
2.2.3 击中不击中变换 8
2.3 结构元素 9
2.4 本章小结 10
3 数学形态学在高分辨率遥感影像道路提取中的应用 11
3.1 高分辨率遥感影像中的道路特征 11
3.2 数学形态学在遥感影像道路提取中的可行性 11
3.2.1 利用阈值分割进行道路提取 11
3.2.2 利用形态学建筑指数（MBI）进行道路提取 14 基于数学形态学的高分辨率遥感影像道路提取:http://www.751com.cn/jisuanji/lunwen_25727.html