Research on Road Centerline Extraction Methods Based on High-resolution Remote Sensing Images
In order to improve the accuracy of road network extraction based on remote sensing images,this paper proposes a road ex-traction 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 ex-tracted 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 net-work in remote sensing images,and the extraction accuracy is better than the comparison algorithms.
万方数据
维普
