Urban Roads Extraction Algorithm For High Resolution Remote Sensing Image

Road is one of the most important artificial features. As the city's backbone, its extraction and updating is of great significance for urban planning, traffic management and so on. In the past few decades, although a lot of solution is put forward to extract road from remote sensing imagery, most of studies have focused on the road extraction low-resolution imagery. Because of the complexity of road characteristics, road extraction in high resolution images is quite differrent from that in low resolution images.Therefore, the study of the urban road network extraction has important theoretical and practical significance.Under this circumstance, this paper researches the urban road extraction of the high resolution remote sensing images based on the deep analysis of the characteristiccs of roads. The main elements are as follows:Firstly, the paper analyzes the characteristic of roads in different resolution images. In combination with previous study, the paper summarys the characteristic of urban road of the high resolution remote sensing images, which is theoretical basis of road extraction. Secondly, a new road segmentation method based on total variations and region growing is presented. Experiments show the new algorithms can seperate the road region from the background. After applying the road segmentation mathod, four typical region geometric factors are selected to detect the typical road segments and remove the non-target segments.Finally, based on the extraction of road area, the paper uses morphology, skeleton extraction and curve fitting to extract the skeleton of the road network.Based on the experiments of several high-resolution images, the experiment results show that the method proposed can extract accurately urban road network from high resolution remote sensing images.