基于高分辨率遥感影像的道路中心线提取方法研究  

Research on Road Centerline Extraction Methods Based on High-resolution Remote Sensing Images

作　　者：张业 徐婧 ZHANG Ye;XU Jing(Ningbo Alatu Digital Technology Co.,Ltd.,Ningbo 315000,China)

机构地区：[1]宁波市阿拉图数字科技有限公司,浙江宁波315000

出　　处：《测绘与空间地理信息》2024年第9期221-224,共4页Geomatics & Spatial Information Technology

摘　　要：为了提高基于遥感影像的道路网络提取精度,本文提出了一种多特征融合的BP神经网络模型的道路提取方法。该方法实现道路提取的步骤为:首先,对原始影像进行色彩变换,提取原始影像光谱饱和度(Satuation,S)信息;其次,提取影像对象的多种特征,将其作为BP神经网络模型的输入,得到初始道路网络并对提取道路结果进行优化;最后,借助道格拉斯-普克(Douglas-Peucker,DP)算法对初始区域轮廓进行多边形逼近并提取道路中心线。通过实测遥感影像数据进行实验,结果表明,本文方法能够准确、完整地提取遥感影像中道路网络,提取精度优于对比算法。In order to improve the accuracy of road network extraction based on remote sensing images,this paper proposes a road extraction method based on multi-feature fusion BP neural network model.The steps of road extraction in this method are as follows:first,color transformation is applied to the original image to extract the spectral saturation(s)information;secondly,various features of image objects are extracted and used as the input of BP neural network model to obtain the initial road network and optimize the extracted road results;finally,with the help of Douglas-Peucker(DP)algorithm,the initial area contour is approximated by polygons and the road centerline is extracted.The experimental results show that this method can accurately and completely extract the road network in remote sensing images,and the extraction accuracy is better than the comparison algorithms.

关 键 词：遥感影像 道路提取 BP神经网络 色彩变换 多特征融合 

分 类 号：P231[天文地球—摄影测量与遥感]