Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Active and passive computed tomography algorithm with a constrained conjugate gradient solution

Conference ·
OSTI ID:3399
An active and passive computed tomographic technique (A&PCT) has been developed at the Lawrence Livermore National Laboratory (LLNL). The technique uses an external radioactive source and active tomography to map the attenuation within a waste drum as a function of mono-energetic gamma-ray energy. Passive tomography is used to localize and identify specific radioactive waste within the same container. The passive data is corrected for attenuation using the active data and this yields a quantitative assay of drum activity. A&PCT involves the development of a detailed system model that combines the data from the active scans with the geometry of the imaging system. Using the system model, iterative optimization techniques are used to reconstruct the image from the passive data. Requirements for high throughput yield measured emission levels in waste barrels that are too low to apply optimization techniques involving the usual Gaussian statistics. In this situation a Poisson distribution, typically used for cases with low counting statistics, is used to create an effective maximum likelihood estimation function. An optimization algorithm, Constrained Conjugate Gradient (CCG), is used to determine a solution for A&PCT quantitative assay. CCG, which was developed at LLNL, has proven to be an efficient and effective optimization method to solve limited-data problems. A detailed explanation of the algorithms used in developing the model and optimization codes is given.
Research Organization:
Lawrence Livermore National Laboratory, Livermore, CA
Sponsoring Organization:
USDOE Office of Defense Programs (DP)
DOE Contract Number:
W-7405-ENG-48
OSTI ID:
3399
Report Number(s):
UCRL-JC-130818; YN0100000; ON: DE00003399
Country of Publication:
United States
Language:
English