A comparative analysis for Haar wavelet efficiency to remove Gaussian and Speckle noise from image
Shallu, Yogendra Narayan, Pankaj Nanglia · 2016
Now a day, the entrance into digital world is making an excrescent growth. So, there is a need to handle vast amount of informative data in efficient and effective manner. But digital information is often contaminated by noise during storage and fetching. In order to make the practical use of data, it is essential to make data noise free. In this context, we describe approximate digital implementation of a mathematical transform, namely wavelet transform. The particular wavelet which is chose here is the Haar wavelet i.e. the simplest wavelet. Here, this transform is applied to denoise standard image corrupted with Speckle and Gaussian noise. The quality of de-noised image has been evaluated on the basis of some factors like Peak Signal to Noise Ratio (PSNR) and Mean Squared Error (MSE). Later, a comparison for efficiency of Haar wavelet using Bay's thresholding in removal of Speckle and Gaussian noise has been carried out where results shows that this wavelet denoising technique is 7.5% efficient in removal of Speckle noise than Gaussian noise.