Conventional reconstruction algorithms for Limited-angle computed tomography introduce blurring and streak artifacts. We propose a reconstruction model that incorporates two regularization terms which play the role of edge-preserving diffusion and smoothing along the x-direction and y-direction respectively. Our model is based on the observation that through the image edges are blurred along the y-direction, they are visible along the x-direction, which is a global view of the theoretical results given by Quinto et al.

This presentation is part of Minisymposium “MS15 - Nonlinear Diffusion: Models, Extensions and Algorithms (2 parts)”
organized by: Ke Chen (University of Liverpool) , Joachim Weickert (Saarland University) , Xue-Cheng Tai (Hong Kong Baptist University) .

Authors:

Peng Zhang (Capital Normal University of Beijing)

Hongwei Li (Capital Normal University of Beijing)

Jinqiu Xu ( Capital Normal University, Beijing)

Yunsong Zhao (Capital Normal University, Beijing)

Keywords:

computed tomography, image reconstruction, inverse problems