The Technology Research Of Urban Road Extraction From High-resolution Remote Sensing Images

In recent years, with the rapid development of remote sensing technology, remote sensing image resolution has improved significantly and high-resolution remote sensing satellites have already been able to provide human with high resolution, large capacity remote sensing image. Extracting target information from remote sensing image has become an important means of updating spatial information, and has been widely used in national economic production and military targets surveillance area. The road network of the remote sensing image is a major source of geographic information, also is the basis of geographic information system data. Extraction of road information on the GIS data acquisition, remote sensing image understanding, mapping and spatial database updates, etc. has far-reaching significance. Therefore, the study of road information extracted from remote sensing image is essential.On the basis of accessing extensive literature, road extraction methods from remote sensing image were studied. We propose two extraction methods, urban road extraction based on the Hough transform and mathematical morphology from remote sensing image, urban road extraction based on the Hough transform and path morphology from remote sensing image. First, pretreatment of remote sensing images; Secondly, binary the image according to the gray scale value; in binary image, based on the Hough transform calculate using the road direction, width and length as constructed structure element of mathematical morphology; Use the basic operators, particle analysis algorithm of mathematical morphology separating road and other objects such as buildings which is connected with road, removing the image spot which is non-road, connecting the broken roads and so on, getting the final extraction results of the road.Experiments show that the two proposed method is feasible and has very good results. Compared with the extraction method based on the Hough transform only and morphology only, extraction method based on the Hough transform and morphology has a better extraction effect.