Local integrated absorbance tomography based on revised iterative reconstruction-reprojection algorithm

摘要

Iterative reconstruction-reprojection (IRR) is an algorithm framework for data extrapolation in missing views of tomography projection. In this paper, this algorithm is revised and applied to reconstruct the distribution of local integrated absorbance in tunable diode laser absorption spectroscopy (TDLAS) tomography system with fixed projection views. The reconstruction process in IRR framework was replaced by algebraic iteration method combined with median filtering to solve the sparse-view reconstruction problem. To validate its effectiveness, the proposed method is compared with the Tikhonov regularization regarding the reconstruction performance based on simulation. It is shown that, when noises present, the revised IRR method has the potential of reconstructing a high spatial resolution image with less artifacts.