Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280491056> ?p ?o ?g. }
- W4280491056 endingPage "104316" @default.
- W4280491056 startingPage "104316" @default.
- W4280491056 abstract "Previous research has shown the high accuracy of convolutional neural networks (CNNs) in asphalt and concrete crack detection in controlled conditions. Yet, human-like generalisation remains a significant challenge for industrial applications where the range of conditions varies significantly. Given the intrinsic biases of CNNs, this paper proposes a vision transformer (ViT)-based framework for crack detection on asphalt and concrete surfaces. With transfer learning and the differentiable intersection over union (IoU) loss function, the encoder-decoder network equipped with ViT could achieve an enhanced real-world crack segmentation performance. Compared to the CNN-based models (DeepLabv3+ and U-Net), TransUNet with a CNN-ViT backbone achieved up to ~61% and ~3.8% better mean IoU on the original images of the respective datasets with very small and multi-scale crack semantics. Moreover, ViT assisted the encoder-decoder network to show a robust performance against various noisy signals where the mean Dice score attained by the CNN-based models significantly dropped (<10%)." @default.
- W4280491056 created "2022-05-22" @default.
- W4280491056 creator A5001529504 @default.
- W4280491056 creator A5017012258 @default.
- W4280491056 creator A5023427388 @default.
- W4280491056 creator A5043170461 @default.
- W4280491056 creator A5060446620 @default.
- W4280491056 creator A5087768372 @default.
- W4280491056 date "2022-08-01" @default.
- W4280491056 modified "2023-10-14" @default.
- W4280491056 title "Vision transformer-based autonomous crack detection on asphalt and concrete surfaces" @default.
- W4280491056 cites W2070902649 @default.
- W4280491056 cites W2331509209 @default.
- W4280491056 cites W2588612844 @default.
- W4280491056 cites W2598457882 @default.
- W4280491056 cites W2889494142 @default.
- W4280491056 cites W2899242765 @default.
- W4280491056 cites W2905163589 @default.
- W4280491056 cites W2912350898 @default.
- W4280491056 cites W2918499589 @default.
- W4280491056 cites W2941356554 @default.
- W4280491056 cites W2943686015 @default.
- W4280491056 cites W2972460946 @default.
- W4280491056 cites W2973071380 @default.
- W4280491056 cites W2986661129 @default.
- W4280491056 cites W2989673213 @default.
- W4280491056 cites W2990423348 @default.
- W4280491056 cites W2997724457 @default.
- W4280491056 cites W3014583121 @default.
- W4280491056 cites W3014945915 @default.
- W4280491056 cites W3016107947 @default.
- W4280491056 cites W3024770686 @default.
- W4280491056 cites W3024912007 @default.
- W4280491056 cites W3040786507 @default.
- W4280491056 cites W3044248863 @default.
- W4280491056 cites W3044580098 @default.
- W4280491056 cites W3045006017 @default.
- W4280491056 cites W3049766255 @default.
- W4280491056 cites W3080584687 @default.
- W4280491056 cites W3085833930 @default.
- W4280491056 cites W3096338035 @default.
- W4280491056 cites W3123663133 @default.
- W4280491056 cites W3128040179 @default.
- W4280491056 cites W3132541064 @default.
- W4280491056 cites W3134108147 @default.
- W4280491056 cites W3136343426 @default.
- W4280491056 cites W3167386507 @default.
- W4280491056 cites W3175962597 @default.
- W4280491056 cites W3182707437 @default.
- W4280491056 cites W3186248524 @default.
- W4280491056 cites W3205511895 @default.
- W4280491056 doi "https://doi.org/10.1016/j.autcon.2022.104316" @default.
- W4280491056 hasPublicationYear "2022" @default.
- W4280491056 type Work @default.
- W4280491056 citedByCount "15" @default.
- W4280491056 countsByYear W42804910562022 @default.
- W4280491056 countsByYear W42804910562023 @default.
- W4280491056 crossrefType "journal-article" @default.
- W4280491056 hasAuthorship W4280491056A5001529504 @default.
- W4280491056 hasAuthorship W4280491056A5017012258 @default.
- W4280491056 hasAuthorship W4280491056A5023427388 @default.
- W4280491056 hasAuthorship W4280491056A5043170461 @default.
- W4280491056 hasAuthorship W4280491056A5060446620 @default.
- W4280491056 hasAuthorship W4280491056A5087768372 @default.
- W4280491056 hasConcept C108583219 @default.
- W4280491056 hasConcept C111919701 @default.
- W4280491056 hasConcept C118505674 @default.
- W4280491056 hasConcept C119599485 @default.
- W4280491056 hasConcept C127413603 @default.
- W4280491056 hasConcept C134306372 @default.
- W4280491056 hasConcept C150899416 @default.
- W4280491056 hasConcept C153180895 @default.
- W4280491056 hasConcept C154945302 @default.
- W4280491056 hasConcept C159985019 @default.
- W4280491056 hasConcept C165801399 @default.
- W4280491056 hasConcept C168056786 @default.
- W4280491056 hasConcept C192562407 @default.
- W4280491056 hasConcept C202615002 @default.
- W4280491056 hasConcept C31972630 @default.
- W4280491056 hasConcept C33923547 @default.
- W4280491056 hasConcept C41008148 @default.
- W4280491056 hasConcept C66322947 @default.
- W4280491056 hasConcept C81363708 @default.
- W4280491056 hasConcept C89600930 @default.
- W4280491056 hasConceptScore W4280491056C108583219 @default.
- W4280491056 hasConceptScore W4280491056C111919701 @default.
- W4280491056 hasConceptScore W4280491056C118505674 @default.
- W4280491056 hasConceptScore W4280491056C119599485 @default.
- W4280491056 hasConceptScore W4280491056C127413603 @default.
- W4280491056 hasConceptScore W4280491056C134306372 @default.
- W4280491056 hasConceptScore W4280491056C150899416 @default.
- W4280491056 hasConceptScore W4280491056C153180895 @default.
- W4280491056 hasConceptScore W4280491056C154945302 @default.
- W4280491056 hasConceptScore W4280491056C159985019 @default.
- W4280491056 hasConceptScore W4280491056C165801399 @default.
- W4280491056 hasConceptScore W4280491056C168056786 @default.
- W4280491056 hasConceptScore W4280491056C192562407 @default.
- W4280491056 hasConceptScore W4280491056C202615002 @default.