An Improved Non-local Mean Ultrasound Image Denoising Algorithm

Abstract

An improved non-local mean ultrasound image denoising method is proposed to solve the problem of poor effect of traditional image denoising algorithm on ultrasound images. In this method the non-local mean filtering algorithm is improved by using the gamma distribution noise model and image block matching, so as to get better noise reduction effect. By comparing with Gaussian filtering algorithm, median filtering algorithm and bilateral filtering algorithm, the improved algorithm achieves better noise reduction than traditional methods, and compared with early non-local mean algorithm in peak signal-to-noise ratio and mean square error performance were optimized.