Markov Chain Monte Carlo Defect Identification in NDE Images Dogandžić, Aleksandar Dogandžić, Aleksandar Zhang, Benhong
dc.contributor.department Electrical and Computer Engineering 2018-02-15T00:11:51.000 2020-06-30T02:03:18Z 2020-06-30T02:03:18Z Mon Jan 01 00:00:00 UTC 2007 2014-10-08 2007-01-01
dc.description.abstract <p>We derive a hierarchical Bayesian method for identifying elliptically‐shaped regions with elevated signal levels in NDE images. We adopt a simple elliptical parametric model for the shape of the defect region and assume that the defect signals within this region are random following a truncated Gaussian distribution. Our truncated‐Gaussian model ensures that the signals within the defect region are higher than the baseline level corresponding to the noise‐only case. We derive a closed‐form expression for the kernel of the posterior probability distribution of the location, shape, and defect‐signal distribution parameters (model parameters). This result is then used to develop Markov chain Monte Carlo (MCMC) algorithms for simulating from the posterior distributions of the model parameters and defect signals. Our MCMC algorithms are applied<em>sequentially</em> to identify multiple potential defect regions. For each potential defect, we construct Bayesian confidence regions for the estimated parameters. Estimated Bayes factors are utilized to rank potential defects (discovered by our sequential scheme) according to goodness of fit. The performance of the proposed methods is demonstrated on experimental ultrasonic <em>C</em>‐scan data from an inspection of a cylindrical titanium billet.</p>
dc.description.comments <p>The following article appeared in <em>AIP Conference Proceedings</em> 894 (2007): 709 and may be found at doi:<a href="" target="_blank">10.1063/1.2718040</a>.</p>
dc.format.mimetype application/pdf
dc.identifier archive/
dc.identifier.articleid 1037
dc.identifier.contextkey 6217048
dc.identifier.s3bucket isulib-bepress-aws-west
dc.identifier.submissionpath ece_pubs/42
dc.language.iso en
dc.source.bitstream archive/|||Sat Jan 15 00:12:08 UTC 2022
dc.source.uri 10.1063/1.2718040
dc.subject.disciplines Applied Statistics
dc.subject.disciplines Electrical and Computer Engineering
dc.subject.disciplines Engineering Physics
dc.subject.disciplines Theory and Algorithms
dc.subject.keywords Bayesian analysis
dc.subject.keywords defect identification
dc.subject.keywords Markov chain Monte Carlo (MCMC)
dc.title Markov Chain Monte Carlo Defect Identification in NDE Images
dc.type article
dc.type.genre conference
dspace.entity.type Publication
relation.isAuthorOfPublication c910f7d3-c386-4c37-8143-4e653a539aa9
relation.isOrgUnitOfPublication a75a044c-d11e-44cd-af4f-dab1d83339ff
Original bundle
Now showing 1 - 1 of 1
850.11 KB
Adobe Portable Document Format