基于高分辨率遥感影像的道路网络提取算法研究

Research on Road Network Extraction Algorithm Based on High-resolution Remote Sensing Image

作为遥感信息分析领域的重要工作之一,高分辨率遥感影像道路提取对于地理信息建库、城市建设规划、城市地形分析与三维表达等方面具有重要的研究意义。本文提出了多种改进算法组成的遥感影像道路网络提取方法,首先使用超像素分割(Simple Linear Iterative Clustering, SLIC)算法对高分辨率遥感影像进行分割处理并使用改进K-means算法对分割后影像进行分类;其次根据绿色植被指数(Green Vegetation Index, CVI)值滤除非人工区域并使用OTSU算法分割提取得到初始道路网络;最后对提取道路网络进行优化处理得到精细化道路网络。使用广州市某地高分辨遥感影像对本文提取算法进行验证,结果表明,本文算法能够有效地提取道路信息,提高道路网络提取精度。

As one of the important work in the field of remote sensing information analysis,high-resolution remote sensing image road extraction has important research significance for geographic information database construction,urban construction planning,urban ter-rain analysis and 3D expression.This paper proposes a road network extraction method of remote sensing image composed of many im-proved algorithms.Firstly,the simple linear iterative clustering(SLIC)algorithm is used to segment the high-resolution remote sens-ing image,and the improved K-means algorithm is used to classify the segmented image;Secondly,the non artificial areas are filtered out according to the green vegetation index(CVI)value,and the initial road network is obtained by Otsu algorithm;Finally,the ex-tracted road network is optimized to obtain a refined road network.A high-resolution remote sensing image of a place in Guangzhou is used to verify the proposed algorithm.The results show that the algorithm can effectively extract road information and improve the ex-traction accuracy of road network.