Abstract: Edge detection is one of the most important imageprocessing steps towards image understanding. Theresultant isolated regions or segments must becompletely separated from their neighbors; that is,edges be continuous. Images must first be smoothed toremove noises. In this paper, a novel fuzzy smootheralgorithm is presented which removes all camera noisesand enhances edge contrasts. A fuzzy edge detectionalgorithm is then presented which is applied on thesmoothed image. In this algorithm normalized hue in HSIand color contrast in RGB spaces are combined using anaggregate operator. Almost local maxima is then foundfor all edge directions and the results are combined.Finding another local maxima produces promisingresults.!13
展开▼
