TV OS-SART with Fractional Order Integral Filtering

Abstract

Computed tomography (CT) is widely used in industrial nondestructive testing (NDT). The projection data of CT usually contain serious noises. In this situation, compared with conventional filtered back projection (FBP), total variation ordered subsets simultaneous algebraic reconstruction technique (TV-OS-SART) is a better reconstruction method. However, image noise will increase after some iterations while the image quality does not meet the requirement. In order to improve the quality of reconstructed image, for three dimensional cone-beam CT, a total variation ordered subsets simultaneous iterative reconstruction algorithm with fractional order integral is researched. Experimental results show that the proposed method has faster convergence speed.