Speckle Noise Reduction from Ultrasound Images using Principal Component Analysis with Bit Plane Slicing and Nonlinear Diffusion method

In this paper we present and evaluate a novel method for an efficient speckle denoising by using principal component analysis (PCA) with bit plane slicing and nonlinear diffusion. We use PCA transformation for generating de-correlated dataset from a noisy image. Then we apply bit plane slicing on the de-correlated dataset and nonlinear diffusion is applied on each bit plane. For nonlinear diffusion in each bit plane level, a gradient threshold is automatically estimated. Add up all bit plane slice after nonlinear diffusion execution and then we implement inverse principal component analysis for making denoised images. The proposed speckle reduction method could improve image quality and the visibility of small structures and fine details in medical ultrasound imaging compared with state-of-the-art speckle denoising algorithms.