Iterative Tomographic Algorithms Of Gas Diffusion Distribution Reconstruction Based On Incomplete Projection Data

Iterative tomographic algorithm is widely used in the reconstruction problem of incomplete projection data. Reconstruction of regional distribution of air pollution through optical remote sensing combined with tomographic techniques is a typical problem of incomplete projection data and fan-beam geometry. Considering the cost of optical instrument, it may not be possible to set a large number of rays to form a complex cross optical path, and the gas diffusion mostly dragging a long tail peak. It is almost impossible for the algebraic iterative algorithm to complete the reconstruction of the distribution of gas diffusion.For the reconstruction of incomplete projection data reconstruction which the gas diffusion distribution facing, numerous research on traditional tomography reconstruction algorithm, especially iterative tomography reconstruction algorithm including algebraic algorithms, and statistical iterative algorithm has been done in this paper. Factors may influence the reconstruction quality have been analyzed further, which provided theoretical basis for algorithm improvement in incomplete projection data. The reconstruction of the Gaussian plume diffusion distribution under the condition of fan beam geometry and double projection direction is deeply discussed in this paper, and the main research work includes:(1)The prior matrix and interval using a smoothing filter strategy are used to improve the classical iterative reconstruction algorithms including ART, MART, MLEM and SART. The improved algorithms have satisfactory reconstruction results of overall peak shape, peak position and peak tail of gas diffusion.(2)A large number of comparative experiments with Gaussian plume model as test model and detailed analysis for effect of ray configuration, relaxation factor, different filter strategies and signal-to-noise ratio on reconstruction results have been carried out. Experimental results show that the algorithms improve the pseudo peak problem, and are robust to noise.(3)Gas diffusion tomography reconstruction GUI software has been accomplished, which integrates the distribution of gas diffusion parameters, ray configurations and the iterative reconstruction algorithm, kinds of reconstruction criterion are also referred. It provides convenience for the analysis of the impact of the gas diffusion distribution factors on the reconstruction, the contrast of reconstruction algorithm and parameter selection.The results can be used as theoretical preparation and algorithm support for the reconstruction of regional distribution of air pollution by optical remote sensing combined with tomographic techniques and then provide evidence for the research on atmospheric transfer, diffusion and settlement process of air pollutants.