In MAXENT 96: Proc. Maximum Entropy Conf., M. Sears, V. Nedeljkovic, N. E. Pendock, and S. Sibisi, eds., pp. 41-51 (Univ. Witwatersrand, Johannesburg, South Africa, 1997)

Uncertainty estimation in reconstructed deformable models

Kenneth M. Hanson, Gregory S. Cunningham, and Robert J. McKee
Los Alamos National Laboratory

Abstract

One of the hallmarks of the Bayesian approach to modeling is the posterior probability, which summarizes all uncertainties regarding the analysis. Using a Markov Chain Monte Carlo (MCMC) technique, it is possible to generate a sequence of objects that represent random samples drawn from the posterior distribution. We demonstrate this technique for a reconstruction of a two-dimensional object from two orthogonal noisy projections. The reconstructed object is modeled in terms of a deformable geometrically-defined boundary with a constant interior density yielding a nonlinear reconstruction problem. We show how an MCMC sequence can be used to estimate uncertainties in the location of the edge of the reconstructed object.

Keywords: > uncertainty estimation, Bayesian estimation, Markov Chain Monte Carlo, deformable geometric model, tomographic reconstruction

Get full paper (pdf, 193 KB)
Return to publication list
Send e-mail to author at kmh@hansonhub.com