An Improved Edge Detection Method based on Median Filter

Abstract

To preserve the edges from corrupted noise, median filter is a popular choice for filtering algorithms. The salt and pepper noise is one of the main problems present while capturing an image. To solve this problem the edges were detected, using different edge detection operators which include Roberts, Sobel, Prewitt and Canny, prior to processing by a median filter. Several simulations were carried out over a number of digital images. The peak signal to noise ratio (PSNR) is higher using Prewitt operator than Sobel operator. Similarly, results also confirm the better performance of Canny edge detection algorithm over others for edge detection however computational cost is high in it compared to other methods.