A New Image Despeckling Method by SRAD Filter and Wavelet Transform Using Bayesian Threshold
Israt Zahan Nishu, Muhammad Fuad Tanjim, A. S. M. Badrudduza, Md. Al Mamun · 2019
Images containing speckle noise are difficult to de-noise since the noise is multiplicative in nature. In this paper an efficient procedure is introduced to effectively eliminate speckle noise from an image and improving the visual quality of the image. The proposed method works with the Speckle Reducing Anisotropic Diffusion (SRAD) in combination with the Discrete Wavelet Transform (DWT) using Bayesian Threshold. In the proposed technique, first the SRAD filter is applied on a log uncompressed noisy image. Then log transformation is performed on the filtered image. Finally, the de-noised image is obtained by performing the DWT using Bayesian thresholding on the log compressed image. The new technique exhibits excellent quantitative and visual performances in terms of Peak-Signal-to-Noise Ratio (PSNR), Root Mean Square Error (RMSE), Structural Similarity Index (SSIM) and Computational Time in seconds over the conventional techniques. The results are demonstrated by simulation on standard test images.