Comparative Analysis of Despeckle Filtering Methods for Ultrasound Images

Abstract

Medical image recognition is becoming more in demand every day. For more accurate and faster diagnosis, it is necessary not only to enhance image recognition systems but also to improve the quality of images. Images taken with ultrasound suffer from speckle noise. Due to the fact that the tissue has a texture, irregularities reflect ultrasonic radiation and create a speckle pattern. Speckles give an understanding of the surface topography, but image distortion takes place. Denoising is a commonly effective way to improve image quality. This paper presents a comparative analysis of popular speckle denoising methods. An anisotropic diffusion filter was created and adapted for this task. This work showed that even such a highly specialized filter has its limitations. A filter was found that can compete with it. The results of this work can be used by scientists and developers as a reference on the main types of filters for removing speckle noise.