Matches in SemOpenAlex for { <https://semopenalex.org/work/W4383901371> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4383901371 endingPage "47" @default.
- W4383901371 startingPage "31" @default.
- W4383901371 abstract "Surface defect detection systems based on deep learning are employed in the manufacturing system, and their good detection performance largely relies on abundant annotated data. Nevertheless, industrial datasets are often difficult to obtain, which hinders the development of defect detection systems to some extent. In an effort to address the issue that the scarcity of training data leads to poor performance, this paper proposes a novel end to end few-shot detection method for industrial real-time detection on aluminum strips. The meta-learning theory is introduced into the multi-scale structure of the YOLOv4 framework, which means that multi-scale meta-feature maps are extracted through the backbone network and each meta-feature map is recalibrated by reweighted vectors to strengthen class-specific features. Subsequently, a multi-scale prediction module with a new parsing strategy is designed to locate and classify the defects at different scales for the reweighted meta-features. All experiments are performed using data collected from the aluminum strip production line in a cold rolling workshop. In addition, the dataset required for training is constructed based on the few-shot training strategy. Pre-training is conducted on a large amount of labeled data from base classes in advance, and the pre-trained weights are then loaded to train the model on a small amount of labeled data from all classes to rapidly deploy the model in multi-task scenarios. The proposed method shows excellent performance, improving the accuracy for novel classes by more than 14.6%, which enhances the industrial practicality of detection systems." @default.
- W4383901371 created "2023-07-12" @default.
- W4383901371 creator A5004981915 @default.
- W4383901371 creator A5024756248 @default.
- W4383901371 creator A5062776702 @default.
- W4383901371 creator A5091650741 @default.
- W4383901371 date "2023-10-01" @default.
- W4383901371 modified "2023-10-01" @default.
- W4383901371 title "Online visual end-to-end detection monitoring on surface defect of aluminum strip under the industrial few-shot condition" @default.
- W4383901371 cites W2024468509 @default.
- W4383901371 cites W2027870374 @default.
- W4383901371 cites W2139963745 @default.
- W4383901371 cites W2782536355 @default.
- W4383901371 cites W2904559969 @default.
- W4383901371 cites W2944303778 @default.
- W4383901371 cites W2963857746 @default.
- W4383901371 cites W2983156430 @default.
- W4383901371 cites W2989368580 @default.
- W4383901371 cites W2997408160 @default.
- W4383901371 cites W2998291476 @default.
- W4383901371 cites W3013328633 @default.
- W4383901371 cites W3024889357 @default.
- W4383901371 cites W3096831136 @default.
- W4383901371 cites W3099928781 @default.
- W4383901371 cites W3127989846 @default.
- W4383901371 cites W3201577723 @default.
- W4383901371 cites W3203678539 @default.
- W4383901371 cites W3217242469 @default.
- W4383901371 cites W4200236189 @default.
- W4383901371 cites W4223460700 @default.
- W4383901371 cites W4280563018 @default.
- W4383901371 cites W4311459800 @default.
- W4383901371 cites W4312737473 @default.
- W4383901371 cites W639708223 @default.
- W4383901371 doi "https://doi.org/10.1016/j.jmsy.2023.06.016" @default.
- W4383901371 hasPublicationYear "2023" @default.
- W4383901371 type Work @default.
- W4383901371 citedByCount "1" @default.
- W4383901371 countsByYear W43839013712023 @default.
- W4383901371 crossrefType "journal-article" @default.
- W4383901371 hasAuthorship W4383901371A5004981915 @default.
- W4383901371 hasAuthorship W4383901371A5024756248 @default.
- W4383901371 hasAuthorship W4383901371A5062776702 @default.
- W4383901371 hasAuthorship W4383901371A5091650741 @default.
- W4383901371 hasConcept C119857082 @default.
- W4383901371 hasConcept C124101348 @default.
- W4383901371 hasConcept C127413603 @default.
- W4383901371 hasConcept C138885662 @default.
- W4383901371 hasConcept C153180895 @default.
- W4383901371 hasConcept C154945302 @default.
- W4383901371 hasConcept C200925200 @default.
- W4383901371 hasConcept C201995342 @default.
- W4383901371 hasConcept C2776401178 @default.
- W4383901371 hasConcept C2780451532 @default.
- W4383901371 hasConcept C41008148 @default.
- W4383901371 hasConcept C41895202 @default.
- W4383901371 hasConcept C74296488 @default.
- W4383901371 hasConceptScore W4383901371C119857082 @default.
- W4383901371 hasConceptScore W4383901371C124101348 @default.
- W4383901371 hasConceptScore W4383901371C127413603 @default.
- W4383901371 hasConceptScore W4383901371C138885662 @default.
- W4383901371 hasConceptScore W4383901371C153180895 @default.
- W4383901371 hasConceptScore W4383901371C154945302 @default.
- W4383901371 hasConceptScore W4383901371C200925200 @default.
- W4383901371 hasConceptScore W4383901371C201995342 @default.
- W4383901371 hasConceptScore W4383901371C2776401178 @default.
- W4383901371 hasConceptScore W4383901371C2780451532 @default.
- W4383901371 hasConceptScore W4383901371C41008148 @default.
- W4383901371 hasConceptScore W4383901371C41895202 @default.
- W4383901371 hasConceptScore W4383901371C74296488 @default.
- W4383901371 hasLocation W43839013711 @default.
- W4383901371 hasOpenAccess W4383901371 @default.
- W4383901371 hasPrimaryLocation W43839013711 @default.
- W4383901371 hasRelatedWork W2382607599 @default.
- W4383901371 hasRelatedWork W2546942002 @default.
- W4383901371 hasRelatedWork W2961085424 @default.
- W4383901371 hasRelatedWork W2970216048 @default.
- W4383901371 hasRelatedWork W3046775127 @default.
- W4383901371 hasRelatedWork W4285260836 @default.
- W4383901371 hasRelatedWork W4286629047 @default.
- W4383901371 hasRelatedWork W4306321456 @default.
- W4383901371 hasRelatedWork W4306674287 @default.
- W4383901371 hasRelatedWork W4224009465 @default.
- W4383901371 hasVolume "70" @default.
- W4383901371 isParatext "false" @default.
- W4383901371 isRetracted "false" @default.
- W4383901371 workType "article" @default.