A data fusion approach to blind deconvolution of multifrequency eddy current data

Mrityunjay Kumar, Pradeep Ramuhalli

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

C-scan data obtained from multifrequency eddy current inspection are smeared representations of true flaws due to finite size of inspection probes. Deconvolution algorithms are commonly used to remove the effect of finite probe size and estimate the true flaw footprint. Single frequency blind deconvolution techniques are insufficient to correctly determine the true flaw footprint since single frequency eddy current inspection does not provide sufficient information about a flaw. This paper proposes a blind deconvolution algorithm that estimates the flaw footprint by using complementary information present across multifrequency eddy current inspection data. Initial results of applying the algorithm to steam generator tubing inspection data indicate the feasibility and robustness of the proposed algorithm.

Original languageEnglish
Title of host publicationReview of Progress in Quantitative Nondestructive Evaluation Volume 24
Pages712-719
Number of pages8
DOIs
StatePublished - Apr 9 2005
Externally publishedYes
EventReview of Progress in Quantitative Nondestructive Evaluation - Golden, CO, United States
Duration: Jul 25 2004Jul 30 2004

Publication series

NameAIP Conference Proceedings
Volume760
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

ConferenceReview of Progress in Quantitative Nondestructive Evaluation
Country/TerritoryUnited States
CityGolden, CO
Period07/25/0407/30/04

Fingerprint

Dive into the research topics of 'A data fusion approach to blind deconvolution of multifrequency eddy current data'. Together they form a unique fingerprint.

Cite this