FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4292347971> ?p ?o ?g. }

• W4292347971 endingPage "12" @default.
• W4292347971 startingPage "1" @default.
• W4292347971 abstract "Surface defect detection is a crucial component in industrial production. Many algorithms based on computer vision have been successfully applied to surface defect detection, but several problems need to be solved in practical implementation. Firstly, surface defects of different products are diverse in shape, size, and location, which pose a challenge to the generalizability and accuracy of the algorithm. Secondly, the algorithm is required to be real-time in practical detection. To address these problems, we propose a multi-scale feature enhancement fusion and reverse attention network for surface defect detection, named MRD-Net, to achieve real-time and end-to-end defect segmentation. The framework first extracts multi-scale feature maps from the pre-trained MobileNetV2. Then the multi-scale feature enhancement fusion module is proposed to enhance and fuse the feature maps of the deeper layers in the backbone to improve detection performance. Through several skip connections, the adjacent branches are connected top-down. The reverse attention module is applied in shallow branches to utilize deeper prediction information. Finally, the boundary refinement module is added to refine the object boundary and improve prediction accuracy. Our proposed method requires only a small number of defective samples and achieves a high detection accuracy. The experimental results on three datasets (14 industrial products) show that the proposed method outperforms the three state-of-the-art segmentation methods in terms of generalizability and accuracy, it also achieves the requirement of real-time detection with a speed of 52 FPS for a 352*352 image." @default.
• W4292347971 created "2022-08-19" @default.
• W4292347971 creator A5012419850 @default.
• W4292347971 creator A5039659323 @default.
• W4292347971 creator A5050029439 @default.
• W4292347971 date "2022-01-01" @default.
• W4292347971 modified "2023-10-18" @default.
• W4292347971 title "MRD-Net: An Effective CNN-Based Segmentation Network for Surface Defect Detection" @default.
• W4292347971 cites W1903029394 @default.
• W4292347971 cites W1915485278 @default.
• W4292347971 cites W1986306729 @default.
• W4292347971 cites W1988963939 @default.
• W4292347971 cites W2012496675 @default.
• W4292347971 cites W2021751319 @default.
• W4292347971 cites W2132083787 @default.
• W4292347971 cites W2145456339 @default.
• W4292347971 cites W2194775991 @default.
• W4292347971 cites W2560023338 @default.
• W4292347971 cites W2598666589 @default.
• W4292347971 cites W2612454721 @default.
• W4292347971 cites W2752782242 @default.
• W4292347971 cites W2772386856 @default.
• W4292347971 cites W2793624836 @default.
• W4292347971 cites W2883780447 @default.
• W4292347971 cites W2884555738 @default.
• W4292347971 cites W2888407265 @default.
• W4292347971 cites W2948982773 @default.
• W4292347971 cites W2954904282 @default.
• W4292347971 cites W2963163009 @default.
• W4292347971 cites W2963881378 @default.
• W4292347971 cites W2964309882 @default.
• W4292347971 cites W2967610771 @default.
• W4292347971 cites W2982083293 @default.
• W4292347971 cites W2990141084 @default.
• W4292347971 cites W3022522270 @default.
• W4292347971 cites W3034713821 @default.
• W4292347971 cites W3083699157 @default.
• W4292347971 cites W3106583357 @default.
• W4292347971 cites W3119769313 @default.
• W4292347971 cites W3127479494 @default.
• W4292347971 cites W3132936317 @default.
• W4292347971 cites W3135243128 @default.
• W4292347971 cites W3146188393 @default.
• W4292347971 cites W3186745756 @default.
• W4292347971 cites W3190244907 @default.
• W4292347971 cites W3208023024 @default.
• W4292347971 cites W3212066961 @default.
• W4292347971 cites W3216537044 @default.
• W4292347971 cites W4210395945 @default.
• W4292347971 cites W4210628166 @default.
• W4292347971 doi "https://doi.org/10.1109/tim.2022.3200361" @default.
• W4292347971 hasPublicationYear "2022" @default.
• W4292347971 type Work @default.
• W4292347971 citedByCount "1" @default.
• W4292347971 countsByYear W42923479712023 @default.
• W4292347971 crossrefType "journal-article" @default.
• W4292347971 hasAuthorship W4292347971A5012419850 @default.
• W4292347971 hasAuthorship W4292347971A5039659323 @default.
• W4292347971 hasAuthorship W4292347971A5050029439 @default.
• W4292347971 hasConcept C105795698 @default.
• W4292347971 hasConcept C119599485 @default.
• W4292347971 hasConcept C124504099 @default.
• W4292347971 hasConcept C127413603 @default.
• W4292347971 hasConcept C134306372 @default.
• W4292347971 hasConcept C138885662 @default.
• W4292347971 hasConcept C141353440 @default.
• W4292347971 hasConcept C153180895 @default.
• W4292347971 hasConcept C154945302 @default.
• W4292347971 hasConcept C27158222 @default.
• W4292347971 hasConcept C2776151529 @default.
• W4292347971 hasConcept C2776401178 @default.
• W4292347971 hasConcept C31258907 @default.
• W4292347971 hasConcept C31972630 @default.
• W4292347971 hasConcept C33923547 @default.
• W4292347971 hasConcept C41008148 @default.
• W4292347971 hasConcept C41895202 @default.
• W4292347971 hasConcept C52622490 @default.
• W4292347971 hasConcept C62354387 @default.
• W4292347971 hasConcept C88796919 @default.
• W4292347971 hasConcept C89600930 @default.
• W4292347971 hasConceptScore W4292347971C105795698 @default.
• W4292347971 hasConceptScore W4292347971C119599485 @default.
• W4292347971 hasConceptScore W4292347971C124504099 @default.
• W4292347971 hasConceptScore W4292347971C127413603 @default.
• W4292347971 hasConceptScore W4292347971C134306372 @default.
• W4292347971 hasConceptScore W4292347971C138885662 @default.
• W4292347971 hasConceptScore W4292347971C141353440 @default.
• W4292347971 hasConceptScore W4292347971C153180895 @default.
• W4292347971 hasConceptScore W4292347971C154945302 @default.
• W4292347971 hasConceptScore W4292347971C27158222 @default.
• W4292347971 hasConceptScore W4292347971C2776151529 @default.
• W4292347971 hasConceptScore W4292347971C2776401178 @default.
• W4292347971 hasConceptScore W4292347971C31258907 @default.
• W4292347971 hasConceptScore W4292347971C31972630 @default.
• W4292347971 hasConceptScore W4292347971C33923547 @default.
• W4292347971 hasConceptScore W4292347971C41008148 @default.
• W4292347971 hasConceptScore W4292347971C41895202 @default.
• W4292347971 hasConceptScore W4292347971C52622490 @default.

• next