FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4282943825> ?p ?o ?g. }

• W4282943825 endingPage "2351" @default.
• W4282943825 startingPage "2339" @default.
• W4282943825 abstract "Deep learning for nondestructive evaluation (NDE) has received a lot of attention in recent years for its potential ability to provide human level data analysis. However, little research into quantifying the uncertainty of its predictions has been done. Uncertainty quantification (UQ) is essential for qualifying NDE inspections and building trust in their predictions. Therefore, this article aims to demonstrate how UQ can best be achieved for deep learning in the context of crack sizing for inline pipe inspection. A convolutional neural network architecture is used to size surface breaking defects from plane wave imaging (PWI) images with two modern UQ methods: deep ensembles and Monte Carlo dropout. The network is trained using PWI images of surface breaking defects simulated with a hybrid finite element / ray-based model. Successful UQ is judged by calibration and anomaly detection, which refer to whether in-domain model error is proportional to uncertainty and if out of training domain data is assigned high uncertainty. Calibration is tested using simulated and experimental images of surface breaking cracks, while anomaly detection is tested using experimental side-drilled holes and simulated embedded cracks. Monte Carlo dropout demonstrates poor uncertainty quantification with little separation between in and out-of-distribution data and a weak linear fit ( <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$R=0.84$ </tex-math></inline-formula> ) between experimental root-mean-square-error and uncertainty. Deep ensembles improve upon Monte Carlo dropout in both calibration ( <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$R=0.95$ </tex-math></inline-formula> ) and anomaly detection. Adding spectral normalization and residual connections to deep ensembles slightly improves calibration ( <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$R=0.98$ </tex-math></inline-formula> ) and significantly improves the reliability of assigning high uncertainty to out-of-distribution samples." @default.
• W4282943825 created "2022-06-16" @default.
• W4282943825 creator A5001793508 @default.
• W4282943825 creator A5036770578 @default.
• W4282943825 creator A5050991438 @default.
• W4282943825 creator A5077306904 @default.
• W4282943825 date "2022-07-01" @default.
• W4282943825 modified "2023-10-16" @default.
• W4282943825 title "Uncertainty Quantification for Deep Learning in Ultrasonic Crack Characterization" @default.
• W4282943825 cites W1534477342 @default.
• W4282943825 cites W1535980446 @default.
• W4282943825 cites W1965555277 @default.
• W4282943825 cites W1981770407 @default.
• W4282943825 cites W2004004400 @default.
• W4282943825 cites W2019328284 @default.
• W4282943825 cites W2047229728 @default.
• W4282943825 cites W2132603731 @default.
• W4282943825 cites W2194775991 @default.
• W4282943825 cites W2334286570 @default.
• W4282943825 cites W2526322432 @default.
• W4282943825 cites W2789664003 @default.
• W4282943825 cites W2794284562 @default.
• W4282943825 cites W2795261237 @default.
• W4282943825 cites W2809788830 @default.
• W4282943825 cites W2895543448 @default.
• W4282943825 cites W2899691698 @default.
• W4282943825 cites W2919115771 @default.
• W4282943825 cites W2930700974 @default.
• W4282943825 cites W2963173190 @default.
• W4282943825 cites W3002369082 @default.
• W4282943825 cites W3003179639 @default.
• W4282943825 cites W3010523353 @default.
• W4282943825 cites W3037946796 @default.
• W4282943825 cites W3098969183 @default.
• W4282943825 cites W3102100346 @default.
• W4282943825 cites W3102888966 @default.
• W4282943825 cites W3105282616 @default.
• W4282943825 cites W3114591654 @default.
• W4282943825 cites W3118577511 @default.
• W4282943825 cites W3148337205 @default.
• W4282943825 cites W3154270017 @default.
• W4282943825 cites W3163239796 @default.
• W4282943825 cites W4230836457 @default.
• W4282943825 doi "https://doi.org/10.1109/tuffc.2022.3176926" @default.
• W4282943825 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35604965" @default.
• W4282943825 hasPublicationYear "2022" @default.
• W4282943825 type Work @default.
• W4282943825 citedByCount "11" @default.
• W4282943825 countsByYear W42829438252022 @default.
• W4282943825 countsByYear W42829438252023 @default.
• W4282943825 crossrefType "journal-article" @default.
• W4282943825 hasAuthorship W4282943825A5001793508 @default.
• W4282943825 hasAuthorship W4282943825A5036770578 @default.
• W4282943825 hasAuthorship W4282943825A5050991438 @default.
• W4282943825 hasAuthorship W4282943825A5077306904 @default.
• W4282943825 hasBestOaLocation W42829438252 @default.
• W4282943825 hasConcept C105795698 @default.
• W4282943825 hasConcept C108583219 @default.
• W4282943825 hasConcept C11413529 @default.
• W4282943825 hasConcept C119857082 @default.
• W4282943825 hasConcept C127313418 @default.
• W4282943825 hasConcept C151730666 @default.
• W4282943825 hasConcept C154945302 @default.
• W4282943825 hasConcept C165838908 @default.
• W4282943825 hasConcept C19499675 @default.
• W4282943825 hasConcept C2776145597 @default.
• W4282943825 hasConcept C2779343474 @default.
• W4282943825 hasConcept C32230216 @default.
• W4282943825 hasConcept C33923547 @default.
• W4282943825 hasConcept C41008148 @default.
• W4282943825 hasConcept C739882 @default.
• W4282943825 hasConcept C81363708 @default.
• W4282943825 hasConceptScore W4282943825C105795698 @default.
• W4282943825 hasConceptScore W4282943825C108583219 @default.
• W4282943825 hasConceptScore W4282943825C11413529 @default.
• W4282943825 hasConceptScore W4282943825C119857082 @default.
• W4282943825 hasConceptScore W4282943825C127313418 @default.
• W4282943825 hasConceptScore W4282943825C151730666 @default.
• W4282943825 hasConceptScore W4282943825C154945302 @default.
• W4282943825 hasConceptScore W4282943825C165838908 @default.
• W4282943825 hasConceptScore W4282943825C19499675 @default.
• W4282943825 hasConceptScore W4282943825C2776145597 @default.
• W4282943825 hasConceptScore W4282943825C2779343474 @default.
• W4282943825 hasConceptScore W4282943825C32230216 @default.
• W4282943825 hasConceptScore W4282943825C33923547 @default.
• W4282943825 hasConceptScore W4282943825C41008148 @default.
• W4282943825 hasConceptScore W4282943825C739882 @default.
• W4282943825 hasConceptScore W4282943825C81363708 @default.
• W4282943825 hasFunder F4320309120 @default.
• W4282943825 hasFunder F4320334627 @default.
• W4282943825 hasIssue "7" @default.
• W4282943825 hasLocation W42829438251 @default.
• W4282943825 hasLocation W42829438252 @default.
• W4282943825 hasLocation W42829438253 @default.
• W4282943825 hasLocation W42829438254 @default.
• W4282943825 hasLocation W42829438255 @default.
• W4282943825 hasOpenAccess W4282943825 @default.
• W4282943825 hasPrimaryLocation W42829438251 @default.

• next