Abstract
This paper presents a hybrid edge detection algorithm in situations where the image is corrupted by Saltand-Pepper noise. Edge detection is an important preprocessing step in image analysis. Successful results of image analysis extremely depend on edge detection. Up to now several edge detection methods have been developed such as Roberts, Prewitt, Sobel, Zero-crossing, Canny, etc. But, they are sensitive to noise. The structure of our proposed edge detector, to make the process robust against noise, is a combination of neural networks, neuro-fuzzy network and adaptive median filter. The internal parameters of these networks are adaptively optimized by training using very simple artificial images that can be generated by computer. The proposed method is tested under noisy conditions on several images and also compared with conventional edge detectors such as Sobel and Canny and a neuro-fuzzy edge detector. Experimental results reveal that the proposed method exhibits better performance and may efficiently be used for the detection of edges in images corrupted by Salt-and-Pepper noise.
