Analysis of Image Denoising using Wavelet Coefficient and Adaptive Subband Thresholding Technique
S. M. Saroja Theerdus Kalavathy, R. M. Suresh · 2011
Image denoising is a common procedure in digital image processing aiming at the removal of noise which may corrupt an image during its acquisition or transmission while sustaining its quality. A statistical model is proposed depending on the magnitude of wavelet coefficients and noise variance for each coefficient is estimated based on the subband it belongs to using Maximum Likelihood (ML) estimator or a Maximum a Posterior (MAP) estimator. An adaptive thresholding is proposed which is applied to each subband coefficient except the low pass or approximation subband. This is done by fixing the optimum thresholding value depending on the decomposition level. The proposed method describes a new method for suppression of noise in image by fusing the wavelet denoising technique with optimized thresholding function to which a multiplying factor (α) is included to make the threshold value dependent on decomposition level. Due to this, the proposed technique yields significantly superior image quality by preserving the edges, producing a better PSNR value. The efficiency is proved on comparing with Bayes shrink (BS), Modified Bayes Shrink (MBS) and Normal Shrink (NS) for different noise level.