Image Denoising by Wavelet Based Thresholding Method

In the present day, visual data transferred in digital imageries is fetching a popular technique of communication, however the picture acquired after transfer communication is frequently distorted by noise. Before it can be used in applications, the received image must be processed. Picture denoising is the process of manipulating picture data in order to create a visibly excellent picture. We can characterize signals with a precise degree of scarcity using wavelet transforms. Wavelet thresholding is a signal estimating approach that uses the wavelet transform's ability to de-noise signals. Noise suppression in medical imaging is a very delicate and challenging endeavour. The trade-off between noise reduction and picture feature preservation must be adjusted in such a way that the diagnostically useful image content is enhanced. The wavelet thresholding method has been widely utilised to de-noise medical images. The goal is to convert the image information into a wavelet basis, where the large coefficients reflect the signal and the smaller coefficients indicate the noise. The noise in the data can be reduced by adjusting these coefficients appropriately. The goal of this thesis is to compare the performance of several thresholding strategies such as Sure Shrink, Visu Shrink, and Bayes Shrink.