FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313654727> ?p ?o ?g. }

• W4313654727 abstract "Convolutional neural networks (CNNs) have achieved tremendous success in pavement crack segmentation. However, it is difficult for CNN-based crack segmentation methods to minimize false-negative and false-positive errors. Compared with false-positive errors, false-negative errors are more difficult to observe and reduce manually. This paper proposes a fine-tuning method for trained CNNs, called policy gradient-based focal loss (focal-PG loss). The trained CNNs will be further trained by focal-PG loss for only one epoch. The proposed focal-PG loss can be applied to reduce the false-negative errors of the trained CNNs by sacrificing their precision. The experimental results show that focal-PG loss greatly improves the crack recognition rate of the trained encoder–decoder network (EDNet). EDNet (focal-PG loss) achieves an overall precision of 96.05%, recall of 99.68%, and F1-score of 97.83% on 100 validation images. In addition, overall precision of 95.53%, recall of 99.58%, and F1-score of 97.51% are observed for the 150 testing images. U-net, LinkNet, and the feature pyramid network are also tested in the paper to validate the effectiveness of focal-PG loss. The results demonstrate that the focal-PG loss can also improve the performance of the aforementioned networks." @default.
• W4313654727 created "2023-01-07" @default.
• W4313654727 creator A5002820983 @default.
• W4313654727 creator A5005864731 @default.
• W4313654727 creator A5006214591 @default.
• W4313654727 creator A5008772277 @default.
• W4313654727 creator A5044296857 @default.
• W4313654727 date "2023-03-01" @default.
• W4313654727 modified "2023-09-29" @default.
• W4313654727 title "Policy Gradient–Based Focal Loss to Reduce False Negative Errors of Convolutional Neural Networks for Pavement Crack Segmentation" @default.
• W4313654727 cites W1901129140 @default.
• W4313654727 cites W2008935780 @default.
• W4313654727 cites W2080610171 @default.
• W4313654727 cites W2101426606 @default.
• W4313654727 cites W2130432690 @default.
• W4313654727 cites W2133059825 @default.
• W4313654727 cites W2560550165 @default.
• W4313654727 cites W2565639579 @default.
• W4313654727 cites W2748643398 @default.
• W4313654727 cites W2767213246 @default.
• W4313654727 cites W2814406141 @default.
• W4313654727 cites W2884786778 @default.
• W4313654727 cites W2886369963 @default.
• W4313654727 cites W2899242765 @default.
• W4313654727 cites W2906977256 @default.
• W4313654727 cites W2912530595 @default.
• W4313654727 cites W2941356554 @default.
• W4313654727 cites W2963351448 @default.
• W4313654727 cites W2963855133 @default.
• W4313654727 cites W2971843891 @default.
• W4313654727 cites W2983902176 @default.
• W4313654727 cites W2989673213 @default.
• W4313654727 cites W2998997213 @default.
• W4313654727 cites W3011200270 @default.
• W4313654727 cites W3022506809 @default.
• W4313654727 cites W3087277009 @default.
• W4313654727 cites W3088602441 @default.
• W4313654727 cites W3102052994 @default.
• W4313654727 cites W3105421998 @default.
• W4313654727 cites W3105636206 @default.
• W4313654727 cites W3135077060 @default.
• W4313654727 cites W3168326173 @default.
• W4313654727 cites W3185554878 @default.
• W4313654727 cites W3199054352 @default.
• W4313654727 cites W3204197025 @default.
• W4313654727 doi "https://doi.org/10.1061/jitse4.iseng-2157" @default.
• W4313654727 hasPublicationYear "2023" @default.
• W4313654727 type Work @default.
• W4313654727 citedByCount "0" @default.
• W4313654727 crossrefType "journal-article" @default.
• W4313654727 hasAuthorship W4313654727A5002820983 @default.
• W4313654727 hasAuthorship W4313654727A5005864731 @default.
• W4313654727 hasAuthorship W4313654727A5006214591 @default.
• W4313654727 hasAuthorship W4313654727A5008772277 @default.
• W4313654727 hasAuthorship W4313654727A5044296857 @default.
• W4313654727 hasConcept C111919701 @default.
• W4313654727 hasConcept C118505674 @default.
• W4313654727 hasConcept C127413603 @default.
• W4313654727 hasConcept C138885662 @default.
• W4313654727 hasConcept C153180895 @default.
• W4313654727 hasConcept C15336307 @default.
• W4313654727 hasConcept C154945302 @default.
• W4313654727 hasConcept C2776401178 @default.
• W4313654727 hasConcept C41008148 @default.
• W4313654727 hasConcept C41895202 @default.
• W4313654727 hasConcept C50644808 @default.
• W4313654727 hasConcept C78762247 @default.
• W4313654727 hasConcept C81363708 @default.
• W4313654727 hasConcept C81669768 @default.
• W4313654727 hasConcept C82552819 @default.
• W4313654727 hasConcept C89600930 @default.
• W4313654727 hasConceptScore W4313654727C111919701 @default.
• W4313654727 hasConceptScore W4313654727C118505674 @default.
• W4313654727 hasConceptScore W4313654727C127413603 @default.
• W4313654727 hasConceptScore W4313654727C138885662 @default.
• W4313654727 hasConceptScore W4313654727C153180895 @default.
• W4313654727 hasConceptScore W4313654727C15336307 @default.
• W4313654727 hasConceptScore W4313654727C154945302 @default.
• W4313654727 hasConceptScore W4313654727C2776401178 @default.
• W4313654727 hasConceptScore W4313654727C41008148 @default.
• W4313654727 hasConceptScore W4313654727C41895202 @default.
• W4313654727 hasConceptScore W4313654727C50644808 @default.
• W4313654727 hasConceptScore W4313654727C78762247 @default.
• W4313654727 hasConceptScore W4313654727C81363708 @default.
• W4313654727 hasConceptScore W4313654727C81669768 @default.
• W4313654727 hasConceptScore W4313654727C82552819 @default.
• W4313654727 hasConceptScore W4313654727C89600930 @default.
• W4313654727 hasIssue "1" @default.
• W4313654727 hasLocation W43136547271 @default.
• W4313654727 hasOpenAccess W4313654727 @default.
• W4313654727 hasPrimaryLocation W43136547271 @default.
• W4313654727 hasRelatedWork W2175746458 @default.
• W4313654727 hasRelatedWork W2732542196 @default.
• W4313654727 hasRelatedWork W2760085659 @default.
• W4313654727 hasRelatedWork W2767651786 @default.
• W4313654727 hasRelatedWork W2769435486 @default.
• W4313654727 hasRelatedWork W3093612317 @default.
• W4313654727 hasRelatedWork W3102253946 @default.
• W4313654727 hasRelatedWork W4200528772 @default.
• W4313654727 hasRelatedWork W4293211451 @default.

• next