Translator Disclaimer
Paper
28 February 2007 Image reconstruction performance as a function of model complexity using information geometry: application to transmission tomographic imaging
Author Affiliations +
Proceedings Volume 6498, Computational Imaging V; 649806 (2007) https://doi.org/10.1117/12.716264
Event: Electronic Imaging 2007, 2007, San Jose, CA, United States
Abstract
Models used to derive image reconstruction algorithms typically make assumptions designed to increase the computational tractability of the algorithms while taking enough account of the physics to achieve desired performance. As the models for the physics become more detailed, the algorithms typically increase in complexity, often due to increases in the number of parameters in the models. When parameters are estimated from measured data and models of increased complexity include those of lower complexity as special cases, then as the number of parameters increases, model errors decrease and estimation errors increase. We adopt an information geometry approach to quantify the loss due to model errors and Fisher information to quantify the loss due to estimation errors. These are unified into one cost function. This approach is detailed in an X-ray transmission tomography problem where allmodels are approximations to the underlying problem defined on the continuum. Computations and simulations demonstrate the approach. The analysis provides tools for determining an appropriate model complexity for a given problem and bounds on information that can be extracted.
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Joseph A. O'Sullivan, Liangjun Xie, David G. Politte, and Bruce R. Whiting "Image reconstruction performance as a function of model complexity using information geometry: application to transmission tomographic imaging", Proc. SPIE 6498, Computational Imaging V, 649806 (28 February 2007); https://doi.org/10.1117/12.716264
PROCEEDINGS
12 PAGES


SHARE
Advertisement
Advertisement
Back to Top