The effect of input limiting on nonlinear filter for the removal of impulsive noise

Mohd Fakru Razi, Bin Mohd Nawi (2009) The effect of input limiting on nonlinear filter for the removal of impulsive noise. [Final Year Project Report] (Unpublished)

	PDF The effect of input limiting on nonlinear filter...(24 pgs).pdf Download (1MB)
	PDF (Please get the password by email to repository@unimas.my , or call ext: 3914 / 3942 / 3933) The effect of input limiting on nonlinear filter...(fulltext).pdf Restricted to Registered users only Download (44MB)

Abstract

Signals are always corrupted by impulsive noise. These corrupted signals are likely to have accidental burst of outlying observations than one would expect from normally distributed signals. This will affect their density functions which are decay in the tails less than Gaussian density function. However, the nonlinear filters provide a useful and flexible approach to face the signals which are contaminated with the impulsive noise. Thus, the aim of this project is to investigate the effects of input limiting on the performance of two-dimensional (2-D) nonlinear filters when the input signal is contaminated by impulsive noise. This project is focused on median filter for filtering purpose and the impulsive noise is modeled as a - stable random process. For simulation purpose, this project was used MATLAB software. The result of simulations showed that the output signal that passed through limited median filter is better in term of their quality.

Actions (For repository members only: login required)

View Item