Denoising Based On Adaptive Fuzzy-weighted Images

This thesis surverys several image denoising methods based on spatial average and fuzzy theory.The strengths and weaknesses of each method is analyzed and commented.In order to remove noise of the corrupted image and maintain the image details simultaneously,our main work is to presents an improved adaptive fuzzy weighted mean algorithm Firstly,we can use a fuzzy rule to determine noisy pixels.Secondly,a new fuzzy membership function can be defined.Finally,we can perform fuzzy smoothing.The numerical experiments are presented to show that the algorithm can remove the noise very efficiently.