Extraction of Roads from Satellite Images Based on Edge Detection

In this paper, an efficient method of extraction of roads from a given set of data base is explained. The extraction of roads plays an important role in urban planning. The other applications of road extractions are identification of isolated buildings that need to be detected and updating of GIS data base according to the requirements of the human expertise. Edge detection techniques such as canny, sobel, and prewitt will be applied to the image. Morphological operations will be applied to the resultant image which consists of noise due to the edge detection operators. The noise present in the resultant image can be reduced by applying Median filtering. The proposed method extracts roads with in less time compared to the earlier methods.