Abstract

Info

Title:: Research progress on limited-angle CT reconstruction algorithm based on total variation model

Author(s):: QI Zeyao; WANG Yuanjun; School of Medical Instrument and Food Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Keywords:: ; Keywords: computed tomography; limited-angle CT reconstruction; total variation; review

Abstract:: Abstract: Using limited-angle CT reconstruction algorithm can reconstruct a CT image of good quality from limited projection data, and the most prominent algorithms among them are based on total variation (TV) model. So far, many related algorithms based on TV model have been proposed to reconstruct images of better quality. Herein the reconstruction algorithms based on TV model are briefly introduced and then studied in two aspects, namely model improvement and solution algorithm. In terms of model improvement, we mainly introduce the methods about how to improve the reconstruction performance of images by introducing relevant prior information (direction information, non-local information, high-order gradient information, etc.) based on the limitations of the TV model and the connotation information of the image. In terms of solution algorithm, we mainly introduce classical gradient descent algorithms and the alternating direction minimization algorithm based on sparse optimization theory. Finally, the existing problems are analyzed and summarized. In conclusion, the combination of model improvement and solution algorithm is still the trend of future development of TV-based limited-angle CT reconstruction algorithm.