An image reconstruction algorithm for spectral exterior CT problem based on edge-preserving diffusion and edge-preserving smoothing

Abstract

The exterior CT problem is a special x-ray CT imaging problem, which is often used for nondestructive inspections of large tubular samples. It can image a pipeline wall with a relatively small detector. However, exterior imaging is challenging owing to the incomplete data obtained. In this paper, we propose a regularization model for the exterior CT problem, in which an edge-preserving diffusion and edge-preserving smoothing regularizer is employed. In addition, the polychromatic characteristic of the emitted x-rays is also considered so that the model is closer to the actual physical process. Using the linearization strategy, a corresponding solution algorithm is derived. The proposed model and the algorithm are verified with both simulated data and real data. It is shown that the two type of artifacts, exterior problem artifacts and beam-hardening artifacts, can be effectively suppressed.