An Iterative probabilistic non-local filter based denoising framework for SAR and Medical images
2019
Image denoising is an essential step in the field of computer vision , medical and remote sensing applications. During the image acquisition, inter and intra variance among the signals that lead to the noisy factor in digital images. In the Synthetic Aperture Radar(SAR) and medical sensors, images are generated with different types of noises such as Gaussian, speckle, impulsive and combined noise. Among these noise types, speckle noise is often a prominent one which reduces the performance of the digital imaging operations. Traditional denoising approaches such as Bayesian denoise, non-local filter, wavelet based shearlet transform, autoencoders etc are used to remove the noise in the speckle noise. These denoising methods are difficult to process ultrasound images and medical images due to the existence of multiple additive, multiplicative and Gaussian noise. Also, these models cannot resolve the issue of sparsity in the low SNR images. To overcome these issues, an Iterative probabilistic non-local filter based denoising technique is proposed to improve the image quality and to minimize the error rate. Proposed approach is effectively denoise the high resolution and low SNR images by using the iterative probabilistic approach .Experimental results are simulated on different realtime noisy images in order to check the effiency of the present approach with the traditional denoising approaches.