Pleiades高分辨率卫星影像道路提取技术

发布时间：2018-03-29 22:08

  本文选题：道路提取　切入点：高分辨率卫星影像　出处：《测绘与空间地理信息》2016年11期

【摘要】：遥感影像上的道路提取是摄影测量与计算机视觉研究的一项重要内容,本文设计了一种高分辨率卫星影像道路提取方法,首先,利用Canny边缘提取算子提取左影像上的边缘,并利用霍夫变换对线特征进行检测,根据道路的几何特性检测平行直线,再基于影像匹配方法匹配立体像对上的道路;然后,利用卫星影像RPC参数前方交会原理实现道路的几何定位。利用Pleiades卫星影像开展试验,自动提取的道路与人工采集的道路几何定位误差在1个像素以内,试验结果验证了本文方法的有效性。
[Abstract]:Road extraction from remote sensing images is an important part of photogrammetry and computer vision research. In this paper, a high resolution satellite image road extraction method is designed. Firstly, Canny edge detection operator is used to extract the edge of the left image. Hough transform is used to detect the line features, parallel lines are detected according to the geometric characteristics of the road, and then the stereo alignment is matched based on the image matching method. Then, The geometric positioning of road is realized by using the principle of intersection of RPC parameters of satellite image, and the geometric positioning error of automatically extracted road and manually collected road is less than 1 pixel by using Pleiades satellite image to carry out the experiment. The experimental results show that the proposed method is effective.
【作者单位】： 河南省测绘地理信息局信息中心;
【基金】：遥感卫星影像道路技术项目(2014AA7033049)资助
【分类号】：P237
，

本文编号：1683063