Prior image constrained low-rank matrix decomposition method in limited-angle reverse helical cone-beam CT.

Abstract

Limited-angle problem occurs in many applications of X-ray computed tomography (CT) and it will lead to artifacts. In this paper, we present the tomographic imaging of pipeline which is in service by cone-beam computed tomography (CBCT). And our purpose is to suppress the artifacts caused by the limited-angle problem. First, traditional helical scanning strategy is adopted for the pipeline where can be fully scanned, and the slices with less or no defects from the reconstructed image are selected to constitute the prior image. Then a limited-angle reverse helical scanning strategy is developed for the pipeline where can be scanned within limited angular range. Second, considering the constraint of the prior image and the resemblance among slices of pipeline, an image model which incorporates the Schatten p(0<p<1)-norm penalty term is developed. Finally, split Bregman algorithm based method is utilized to solve the model. Numerical results show that our method is efficient and feasible in suppressing artifacts and it is superior to popular methods when the angular range is small. The developed method is promising in suppressing the artifacts in limited-angle problems, and it can be potentially used in pipeline testing.