Bayesian image reconstruction using image‐modeling Gibbs priors

Chan, M.; Herman, Gabor T.; Levitan, E.

doi:10.1002/(sici)1098-1098(1998)9:2/3<85::aid-ima4>3.3.co;2-k

Cited by 12 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…trust-region methods for nonlinear minimization (PCG, quasi-Newton methods) [6,14], stochastic sampling (Metropolis-Hasting or Gibbs sampling) [12,43], pattern search methods (Poll method), large-scale primal-dual interior-point methods (e.g. [44]), or D.C. algorithm [47] with convex-concave regularization [55,56].…”

Section: Methodsmentioning

confidence: 99%

On image reconstruction algorithms for binary electromagnetic geotomography

Zdunek

2008

Theoretical Computer Science

View full text Add to dashboard Cite

a b s t r a c tIn this paper we discuss the selected image reconstruction methods of binary tomography in the context of their application to geophysical imaging. We restrict our considerations to a discrete version of high-frequency electromagnetic geotomography, which we label as Binary Electromagnetic Geotomography (BEG). Basically, such an imaging technique may be applied to detect subsurface anomalies (air-filled voids) whose attenuation coefficient is very low (nearly zero-value) and considerably different from that for the background.The assumption for a binary representation of the image to be reconstructed substantially relaxes image reconstruction problems related to ill-posedness that comes from an intrinsic limitation of an angular range of projections. We test two algorithms for binary tomography, where the penalty term is based on the Markov Random Field (MRF) model. The mean-field reference distribution and mean-field annealing are applied to estimate the global maximizer of the Gibbs-Boltzmann distribution associated with the objective function. We also apply the projected gradient algorithm that uses a binary steering. Very efficient implementations of the algorithms are also given. The numerical results are presented for noise-free, noisy, and real data.

show abstract

Section: Methodsmentioning

confidence: 99%

On image reconstruction algorithms for binary electromagnetic geotomography

Zdunek

2008

Theoretical Computer Science

View full text Add to dashboard Cite

show abstract

“…Actually, the more transducers used, the more accurate the measurement, but the additional machining, mounting, and instrumentation cost and possible cosmetic problems will be introduced and pose a problem prohibiting its application. Although a continuous image of pressure profile cannot be acquired directly due to the limited number of pressure sensors, one particular Markov Chain Monte Carlo (MCMC) method, the Gibbs sampling, can be utilized to cope with incomplete information (Chan et al 1998;Geman and Geman 1984;Givens and Hoeting 2005;Martinez and Martinez 2002). It is well suited here to construct the continuous image based on the limited observed information.…”

Section: Measurement Of Pressure Profilementioning

confidence: 99%

A Finite Element-based Heuristic Estimation of Local Preform Permeability for Resin Transfer Molding

Qiu

Zhang

et al. 2008

Transp Porous Med

View full text Add to dashboard Cite

Uniformity of fabrics significantly affects the resin flow behavior in the resin transfer molding (RTM) process. Due to fabric defects or improper fiber preform preparation/loading, non-uniformity in fabric structure frequently occurs in RTM processing and creates local permeability variations. Such variations often lead to unbalanced resin flow patterns and thus result in defects of finished composite parts. In RTM process modeling, an accurate estimation of the whole field permeability profile of the fiber preform is critical for predicting resin flow pattern correctly. In this article, a finite element-based heuristic computing method is introduced for estimating the in situ whole-field isotropic permeability profile of the preform using a steady flow of gas. Compared with conventional approaches, this method is effective in measuring local permeability variations and applicable to molds with complex 2-D geometries, as well as diverse injection strategies. Several case studies were presented with experimental designs and numerical computations to illustrate the effectiveness and efficiency of the method.

show abstract

“…However, in (Geman and Geman, 1984;Derin and Cole, 1986;Lakshmanan and Derin, 1989) the segmentation performance is limited by the fact that the pairwise Gibbs prior model contains only incomplete boundary information. In (Chan et al, 1997(Chan et al, , 1998, Chan used a high order Gibbs prior model which integrates the region and the boundary information to reconstruct medical images from 2D projections. The advantage of this approach is to use a Gibbs distribution that integrates the boundary information instead of previous Gibbs prior models that only considers the pairwise pixel similarity.…”

Section: Previous Work On Gibbs Prior Modelsmentioning

confidence: 99%