Performance Analysis of Median and Wiener Filters in Image Denoising

Abstract

Images are likely to be degraded by the sensing environment when acquired through optical, electro optical or electronic means. The degradation may be in the form of sensor noise, blur due to camera misfocus, relative object-camera motion, random atmospheric turbulence, etc. For digitally acquired pictures, noise can be summarized as the visible effects of an electronic error in the final image. Noise is a function of how well the image sensor and digital signal processing systems inside a digital camera are prone to and can cope with or remove these errors. Noise significantly degrades the image quality and increases the difficulty in discriminating fine details in the image. It also complicates further image understanding and low-level computer vision processing, such as image segmentation and edge detection. This work aims to understand the performance of various filters on different noises. The image under test was subjected to Gaussian noise, salt & pepper noise and Speckle noise to two input image samples. They were further analysed with respect to the performance of Median filter, Wiener filter and Order statistic filter in denoising the images. The experiments proved and concluded that Median filter is capable of removing salt & pepper noise in images to a greater extent, while other filters used in this work could not remove the same. In this work the parameter SNR (signal to noise ratio) was used to decide the effectiveness of filter. The SNR value for the salt and pepper noise after applying the median filter gave as high as 23, which proves the fact that median filter is most suitable to remove the salt and pepper noise.