Analysis of Edge Detection Algorithms and Denoising Filters on Digital X-Ray Images

Abstract

Image Enhancement is one of the preprocessing steps that have to be performed in all images for enhancing the given image in order to improve the quality of that image. One of the most important tasks in image processing is the removal of noise from an image. The greatest challenge is that while denoising, edges of images should be preserved and originality of image must not be compromise. In this paper we made an attempt to analyze performance of various filters over different types of noise and evaluate performance of various edge detecting algorithms. Digital x-ray image is taken as input and various noises are added to it. For each kind of noise (Salt & pepper, Gaussian, Speckle and Poisson) different types of filters (Mean, Median, Weiner and Gaussian) are applied and the parameters such as Mean Square Error, Average Difference and Structural Content are measured. Next, the input image is imposed on various edge detection algorithms and then comparison is done. Time taken by each algorithm (Canny, Sobel, Log, Roberts and Prewitt) to detect edge is measured in seconds. Finally suitable denoising filter and edge detecting algorithm are identified for digital x-ray image.