Authors : P. Thirumurugan and K. Veerakumar
Abstract: Medical images are often corrupted by impulse noises during the process of medical image acquisition and transmission. Hence, an efficient denoising technique is very important for the medical image processing applications before diagnosing. In this study, Euclidean detector noise filtering algorithm is proposed which detects and removes the impulse noise from the noise affected images, especially while compared with other state-of-the-art methods such as a switched median filter and selective median filter interms of medical image quality and mean square error. The process of impulse noise detection includes finding of the Euclidean direction, through the estimation of standard deviation in various directions of the filtering window. It also preserves the edges in the medical images during the denoising process. The main contribution of this study is to detect and reduce the impulse noises from the medical images for further medical diagnosis. The proposed method has been tested using different modalities of 20 medical images and proved to provide a better visual quality when compared to conventional methods.
P. Thirumurugan and K. Veerakumar, 2016. Analysis of Euclidean Denoising Algorithm for Medical Image Modalities. Asian Journal of Information Technology, 15: 184-190.