Matches in SemOpenAlex for { <https://semopenalex.org/work/W3080355920> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3080355920 endingPage "4849" @default.
- W3080355920 startingPage "4849" @default.
- W3080355920 abstract "The excellent generalization ability of deep learning methods, e.g., convolutional neural networks (CNNs), depends on a large amount of training data, which is difficult to obtain in industrial practices. Data augmentation is regarded commonly as an effective strategy to address this problem. In this paper, we attempt to construct a crack detector based on CNN with twenty images via a two-stage data augmentation method. In detail, nine data augmentation methods are compared for crack detection in the model training, respectively. As a result, the rotation method outperforms these methods for augmentation, and by an in-depth exploration of the rotation method, the performance of the detector is further improved. Furthermore, data augmentation is also applied in the inference process to improve the recall of trained models. The identical object has more chances to be detected in the series of augmented images. This trick is essentially a performance–resource trade-off. For more improvement with limited resources, the greedy algorithm is adopted for searching a better combination of data augmentation. The results show that the crack detectors trained on the small dataset are significantly improved via the proposed two-stage data augmentation. Specifically, using 20 images for training, recall in detecting the cracks achieves 96% and Fext(0.8), which is a variant of F-score for crack detection, achieves 91.18%." @default.
- W3080355920 created "2020-09-01" @default.
- W3080355920 creator A5016942883 @default.
- W3080355920 creator A5034526233 @default.
- W3080355920 creator A5042354869 @default.
- W3080355920 creator A5042981412 @default.
- W3080355920 date "2020-08-27" @default.
- W3080355920 modified "2023-10-16" @default.
- W3080355920 title "CNN Training with Twenty Samples for Crack Detection via Data Augmentation" @default.
- W3080355920 cites W2005029343 @default.
- W3080355920 cites W2031489346 @default.
- W3080355920 cites W2055122019 @default.
- W3080355920 cites W2055755574 @default.
- W3080355920 cites W2598457882 @default.
- W3080355920 cites W2730181302 @default.
- W3080355920 cites W2751134959 @default.
- W3080355920 cites W2793387000 @default.
- W3080355920 cites W2897772777 @default.
- W3080355920 cites W2986188726 @default.
- W3080355920 cites W2996624809 @default.
- W3080355920 cites W3016107947 @default.
- W3080355920 doi "https://doi.org/10.3390/s20174849" @default.
- W3080355920 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7506713" @default.
- W3080355920 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/32867223" @default.
- W3080355920 hasPublicationYear "2020" @default.
- W3080355920 type Work @default.
- W3080355920 sameAs 3080355920 @default.
- W3080355920 citedByCount "17" @default.
- W3080355920 countsByYear W30803559202021 @default.
- W3080355920 countsByYear W30803559202022 @default.
- W3080355920 countsByYear W30803559202023 @default.
- W3080355920 crossrefType "journal-article" @default.
- W3080355920 hasAuthorship W3080355920A5016942883 @default.
- W3080355920 hasAuthorship W3080355920A5034526233 @default.
- W3080355920 hasAuthorship W3080355920A5042354869 @default.
- W3080355920 hasAuthorship W3080355920A5042981412 @default.
- W3080355920 hasBestOaLocation W30803559201 @default.
- W3080355920 hasConcept C100660578 @default.
- W3080355920 hasConcept C108583219 @default.
- W3080355920 hasConcept C111919701 @default.
- W3080355920 hasConcept C119857082 @default.
- W3080355920 hasConcept C134306372 @default.
- W3080355920 hasConcept C138885662 @default.
- W3080355920 hasConcept C153180895 @default.
- W3080355920 hasConcept C154945302 @default.
- W3080355920 hasConcept C177148314 @default.
- W3080355920 hasConcept C2776214188 @default.
- W3080355920 hasConcept C33923547 @default.
- W3080355920 hasConcept C41008148 @default.
- W3080355920 hasConcept C41895202 @default.
- W3080355920 hasConcept C74050887 @default.
- W3080355920 hasConcept C76155785 @default.
- W3080355920 hasConcept C81363708 @default.
- W3080355920 hasConcept C94915269 @default.
- W3080355920 hasConcept C98045186 @default.
- W3080355920 hasConceptScore W3080355920C100660578 @default.
- W3080355920 hasConceptScore W3080355920C108583219 @default.
- W3080355920 hasConceptScore W3080355920C111919701 @default.
- W3080355920 hasConceptScore W3080355920C119857082 @default.
- W3080355920 hasConceptScore W3080355920C134306372 @default.
- W3080355920 hasConceptScore W3080355920C138885662 @default.
- W3080355920 hasConceptScore W3080355920C153180895 @default.
- W3080355920 hasConceptScore W3080355920C154945302 @default.
- W3080355920 hasConceptScore W3080355920C177148314 @default.
- W3080355920 hasConceptScore W3080355920C2776214188 @default.
- W3080355920 hasConceptScore W3080355920C33923547 @default.
- W3080355920 hasConceptScore W3080355920C41008148 @default.
- W3080355920 hasConceptScore W3080355920C41895202 @default.
- W3080355920 hasConceptScore W3080355920C74050887 @default.
- W3080355920 hasConceptScore W3080355920C76155785 @default.
- W3080355920 hasConceptScore W3080355920C81363708 @default.
- W3080355920 hasConceptScore W3080355920C94915269 @default.
- W3080355920 hasConceptScore W3080355920C98045186 @default.
- W3080355920 hasFunder F4320336213 @default.
- W3080355920 hasIssue "17" @default.
- W3080355920 hasLocation W30803559201 @default.
- W3080355920 hasLocation W30803559202 @default.
- W3080355920 hasLocation W30803559203 @default.
- W3080355920 hasOpenAccess W3080355920 @default.
- W3080355920 hasPrimaryLocation W30803559201 @default.
- W3080355920 hasRelatedWork W2731899572 @default.
- W3080355920 hasRelatedWork W2999805992 @default.
- W3080355920 hasRelatedWork W3116150086 @default.
- W3080355920 hasRelatedWork W3133861977 @default.
- W3080355920 hasRelatedWork W4200173597 @default.
- W3080355920 hasRelatedWork W4223943233 @default.
- W3080355920 hasRelatedWork W4291897433 @default.
- W3080355920 hasRelatedWork W4312417841 @default.
- W3080355920 hasRelatedWork W4321369474 @default.
- W3080355920 hasRelatedWork W4380075502 @default.
- W3080355920 hasVolume "20" @default.
- W3080355920 isParatext "false" @default.
- W3080355920 isRetracted "false" @default.
- W3080355920 magId "3080355920" @default.
- W3080355920 workType "article" @default.