Design of Denoising Algorithm for Removal of Impulse Noise in Images

A new framework proposes to remove the high density impulse noise from the digital images using spatial filtering techniques. Here, the performance will be compared between decision based un symmetric trimmed median filter and previous non linear filters. In the Transmission of images over channels, Images are corrupted by impulse noise, due to faulty communications. Impulse noise is also referred to as Salt and pepper noise. The aim of filtering is to remove the impulses so that the noise free image is fully recovered with minimum signal distortion. The best and most widely used non-linear digital filters are median filters. Median filters are known for their potential to remove impulse noise without damaging the edges. Adaptive Median is a "decision-based" or "switching" filter that first identifies possible noisy pixels and then replaces them using the median filter, while leaving all other pixels unchanged. This filter is excellent at detecting noise even at a high noise level. The adaptive structure of this filter fortify that the impulse noises are detected even at a high noise level provided that the Window size is large enough. The accessible non-linear filter like Standard Median Filter (SMF), Adaptive Median Filter (AMF), Decision Based Algorithm (DBA) and Robust Estimation Algorithm (REA) shows better results at low and medium noise densities. At high noise density, their performance is deprived. A new algorithm to remove high-density impulse noise using Decision Based UN Symmetric Trimmed Median Filter (DBUTM) is used.