New switching based order statistic Filters for the removal of impulse Noise in images

Abstract

Images and videos are often corrupted by noise and degradations newlinedue to faulty communications or noisy channels or faulty camera in the newlinetransmission of images and videos over channels Noise removal is one of the newlinemost important and challenging tasks in signal and image processing newlineIn images high frequency components carry valuable information like newlineedges and fine details In the presence of impulse noise reasonable noise newlinesuppression can be achieved by nonlinear filters where linear filters will newlinesmear the edges while smoothing noise Linear filters fail in preserving edges newlinein images corrupted by non Gaussian noise Nonlinear filters are effective in newlinethe removal of non Gaussian noise with edge and fine detail preservation newlineOne of the most popular nonlinear filters for impulse noise removal is order newlinestatistic filters There are large numbers of nonlinear filters and their newlineextensions are under research The most commonly used order statistic filter newlineis median filter since it is easy to implement and removes impulse noise while newlinepreserving edges This thesis introduces some nonlinear filtering schemes newlinebased on order statistics New median based algorithms are proposed to newlineaddress problems in image denoising such as detection and removal of newlineimpulse noise without disturbing edges and fine details newline