Concluding Report: Quantitative Tomography Simulations and Reconstruction Algorithms

Aufderheide, M B ; Martz, H E ; Slone, D M ; Jackson, J A ; Schach von Wittenau, A E ; Goodman, D M ; Logan, C M ; Hall, J M

Lawrence Livermore National Laboratory

关键词: Simulation;    Optimization;    Algorithms;    71 Classical And Quantum Mechanics, General Physics;    Tomography;   

DOI  :  10.2172/15002511
RP-ID  :  UCRL-ID-146938
RP-ID  :  W-7405-ENG-48
RP-ID  :  15002511

美国|英语

来源: UNT Digital Library

【 摘 要 】

In this report we describe the original goals and final achievements of this Laboratory Directed Research and Development project. The Quantitative was Tomography Simulations and Reconstruction Algorithms project (99-ERD-015) funded as a multi-directorate, three-year effort to advance the state of the art in radiographic simulation and tomographic reconstruction by improving simulation and including this simulation in the tomographic reconstruction process. Goals were to improve the accuracy of radiographic simulation, and to couple advanced radiographic simulation tools with a robust, many-variable optimization algorithm. In this project, we were able to demonstrate accuracy in X-Ray simulation at the 2% level, which is an improvement of roughly a factor of 5 in accuracy, and we have successfully coupled our simulation tools with the CCG (Constrained Conjugate Gradient) optimization algorithm, allowing reconstructions that include spectral effects and blurring in the reconstructions. Another result of the project was the assembly of a low-scatter X-Ray imaging facility for use in nondestructive evaluation applications. We conclude with a discussion of future work.