Low Dose CT Image De-Noising Using BM3D Filtering Algorithms

Abstract

Computed tomography (CT) scan is one of the important scanning systems in the medical field. It uses the computer-processed combinations of many X-ray measurements obtained from many angles which produce the  images of specific cross sectional part of scanned object. It helps users to see the interior of the object and it does not make a cut. In CT- scanning process, there must be chance to produce radiation effect. So low dose techniques are used to reduce this radiation effect. Low dose CT images can be corrupted by this type of radiations. Hence, we had proposed a block matching filtering approach to reduce this corruption in the low dose CT images. The presented filtering algorithm can be modified based on the color filter array and alpha rooting approaches. The performance of the presented algorithm can be validated by PSNR, SSIM, DSSIM, MSE and MAE measurement. The overall implementation is held in Matlabr2014a version.