| Computed Tomography imaging technology has been widely used in the field of medical diagnosis and industrial test. using X-ray imaging, but will produce a certain amount of radiation damage, could potentially increases the risk of causing cancer, thus we need to reduce the X-ray radiation dose as much as possible on the premise of guarantee the quality of CT image.Sparse view projection reconstruction method Based on theory of compressed sensing can effectively reduce the radiation dose.This article utilize SART algorithm as the data fidelity term, and regularization term as the constraint penalty term, reach the aim of effectively improving the quality of restructured image.While the accuracy and speed of solution for the regularized minimization sub-problem is crucial for the great quality and less solving time of restructured image.This article holds in-depth discussions on CT sparse-view projection regularization optimization method based on Split-Bregman iterative technique, the main work is divided into the following points:(1) Aiming at the potential problems in the ART-TV algorithm, TV minimization problem shows slow rate of convergence, and TV regularization often produce cartoon-like approximations, thus need to find other more suitable for regular constraint penalty term. This article studies general iteration solutions based on Split-Bregman iterative sparse reconstruction regularization framework, in addition to have a better analysis and research of the performance of various regularization terms.(2) In light of the regularization function selection problem, including Total Variation(TV), Non-Local Total Variation(NLTV) and Bilateral Total Variation(BTV)regularization models, Analysis the present regularization methods about their defects,take advantage of1? norm and2? norm, propose a regularization called BTV regularization method with norm adaptation(NABTV) under the proposed framework.Experimental results proves the validity of this proposed method. |
PDF Full Text Request