融合纹理和形状特征的高分辨率遥感影像道路提取

提出一种结合纹理和形状特征提取道路信息的方法。首先利用灰度共生矩阵提取纹理特征,并将其应用于最大似然分类中提取面状道路,然后利用形态学方法分割道路与其相连地物,最后利用提出的3个形状指数(凹度、精密度、偏心角)有效地识别和区分了道路与非道路地物,并最终实现了提纯道路的目的。实验结果证明,该方法可以准确地提取主干道路网,剔除非道路地物的影响。

Road Extraction from High-resolution Remote Sensing Images Based on Texture and Shape Features

As one of the most important features in geographic database,road extraction is always the research focus in the field of remote sensing.This paper proposes a new method integrating texture and shape features for road extraction.Firstly,the texture feature obtained by gray-level co-occurrence matrix（GLCM） is applied to maximum likelihood classification and to extract the road surface image;then the morphological methods are utilized to segment the road and non-road objects;finally,three shape features are presented to refine the road information and eliminate the influence of non-road objects,such as buildings and parking lots.Experimental results indicate that this method is efficient to extract the central road network accurately.