A non-local total generalized variation regularization reconstruction method for sparse-view x-ray CT

Abstract

Sparse-view X-ray computed tomography (CT) reconstruction, employing total generalised variation (TGV), effectively mitigates the stepwise artefacts associated with total variation (TV) regularisation while preserving structural features within transitional regions of the reconstructed image. Despite TGV surpassing TV in reconstruction quality, it neglects the non-local self-similarity prior, recognised for its efficacy in restoring details during CT reconstruction. This study introduces a non-local total generalised variation (NLTGV) to address the limitation of TGV regularisation method. Specifically, we propose an NLTGV-regularised method for sparse-view CT reconstruction, utilising non-local high-order derivative information to maintain image features and non-local self-similarity for detail recovery. Owing to the non-differentiability of the NLTGV regulariser, we employ an alternating direction method of multipliers (ADMM) optimisation method, facilitating an efficient solution by decomposing the reconstruction model into sub-problems. The proposed method undergoes evaluation using both simulated and real-world projection data. Simulation and experimental results demonstrate the efficacy of the proposed approach in enhancing the quality of reconstructed images compared to other competitive variational reconstruction methods. In conclusion, the simultaneous incorporation of sparsity priors of high-order TV and non-local similarity proves advantageous for structural detail recovery in sparse-view CT reconstruction.

Low-resolution prior equilibrium network for CT reconstruction

“…Different kinds of prior knowledge have been investigated for incomplete data reconstruction problems. The non-local similarity has been shown highly beneficial for enhancing fine details in reconstructed CT images [21,32,35,45]. Both low-rank prior and dictionary learning have been used to leverage the sparsity of the image in a transform domain [11,41,53].…”