Image Denoising Using Hybrid Thresholding, DWT and Adaptive Intensity Transformations

Although various denoising algorithms have been proposed, we all know for a fact that still removing noise from images is a great issue of concern for researchers. The proposed mechanisms have not been able to attain the desirable results. The noise gets introduced while acquisition or transmission of the image. One of the majorly found noise in images is the Gaussian noise. The corruption of image with Gaussian noise is a classical problem till date. In this paper I have proposed a new method for the removal of Gaussian noise. The technique is majorly based on a hybrid of Neigh and Bayes shrink, and Discrete Wavelet Transform. This technique is a kind of extension of previous techniques. The main aim is to minimize the noise as much as possible. The results have been compared on various quality parameters such as PSNR, CNR, Standard Deviation and Entropy.