This paper aimed segmenting the road network from aerial images. Fuzzy mathematical morphology has been proposed as a new method of image processing, especially in the analysis of features from an ambiguous image. Fuzzy morphological operators work with two images: an original to be processed and a structuring element . In this paper we have developed an algorithm for segmentation of a remotely sensed image to extract the urban road network using Fuzzy Mathematical Morphology. In the vast field of image processing , Fuzzy Mathematical Morphology is not only an established image analysis technique with applications in all disciplines dealing with digital spatial data but also alternative extension of binary morphology to gray scale morphology using techniques from fuzzy set theory and fuzzy logic. It allows processing images to enhance fuzzy areas, segment objects, detect edges and analyze the structures very clearly. The algorithm developed in this paper is a novel segmentation method that segments the road from the satellite aerial images with less computational complexity compared to traditional Mathematical Morphology. 

(P1150825001, 2.89 MB)