Matches in SemOpenAlex for { <https://semopenalex.org/work/W4283523645> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4283523645 endingPage "1616" @default.
- W4283523645 startingPage "1600" @default.
- W4283523645 abstract "In industrial scenarios, knowledge of the severity of defects is needed to distinguish different levels of defects and adapt to the possible modification of defect identification criteria. Well-calibrated detection models are crucial, especially for these decision-making applications. Previous works have measured and calibrated biased confidence estimates of classification and object detection models, while the field of severity calibration has not yet been addressed. Through experiments, we find that the severity also has a specific relationship with the probabilistic prediction of models. Therefore, we present a framework to measure and calibrate severity calibration for defect detection applications by constructing maps between probabilistic prediction, precision, and severity. We find that unlike confidence calibration, the reliability diagram of a well-calibrated model for severity calibration is funnel-shaped instead of linear-shaped. To approximate the funnel-shaped mapping function, we introduce a Gaussian distribution -based function with parameters that are fitted by marked data. Additionally, extending algorithms are proposed to adjust classical confidence calibration methods (e.g., histogram binning, isotonic regression, vector scaling, and temperature scaling) for severity calibration. By mapping the probabilistic prediction to precision, severity calibration is transformed to confidence calibration. The proposed severity calibration methods are evaluated by the expected severity calibration error (ESCE) in comparative experiments, which show that the extended isotonic regression achieves the best performance. In Zhongce’s 21 intelligent detection lines, we applied severity calibration and received good feedback from inspectors." @default.
- W4283523645 created "2022-06-27" @default.
- W4283523645 creator A5028948842 @default.
- W4283523645 creator A5034790811 @default.
- W4283523645 creator A5060002817 @default.
- W4283523645 creator A5069353502 @default.
- W4283523645 creator A5069969191 @default.
- W4283523645 date "2022-08-01" @default.
- W4283523645 modified "2023-09-27" @default.
- W4283523645 title "A novel severity calibration algorithm for defect detection by constructing maps" @default.
- W4283523645 cites W1540419678 @default.
- W4283523645 cites W1861492603 @default.
- W4283523645 cites W2031489346 @default.
- W4283523645 cites W2098824882 @default.
- W4283523645 cites W2137556846 @default.
- W4283523645 cites W2905416267 @default.
- W4283523645 cites W2926743806 @default.
- W4283523645 cites W2964241181 @default.
- W4283523645 cites W2980587280 @default.
- W4283523645 cites W3009635072 @default.
- W4283523645 cites W3046756309 @default.
- W4283523645 cites W3092482776 @default.
- W4283523645 doi "https://doi.org/10.1016/j.ins.2022.06.076" @default.
- W4283523645 hasPublicationYear "2022" @default.
- W4283523645 type Work @default.
- W4283523645 citedByCount "3" @default.
- W4283523645 countsByYear W42835236452023 @default.
- W4283523645 crossrefType "journal-article" @default.
- W4283523645 hasAuthorship W4283523645A5028948842 @default.
- W4283523645 hasAuthorship W4283523645A5034790811 @default.
- W4283523645 hasAuthorship W4283523645A5060002817 @default.
- W4283523645 hasAuthorship W4283523645A5069353502 @default.
- W4283523645 hasAuthorship W4283523645A5069969191 @default.
- W4283523645 hasConcept C105795698 @default.
- W4283523645 hasConcept C11413529 @default.
- W4283523645 hasConcept C115961682 @default.
- W4283523645 hasConcept C121332964 @default.
- W4283523645 hasConcept C124101348 @default.
- W4283523645 hasConcept C153180895 @default.
- W4283523645 hasConcept C154945302 @default.
- W4283523645 hasConcept C163258240 @default.
- W4283523645 hasConcept C165838908 @default.
- W4283523645 hasConcept C33923547 @default.
- W4283523645 hasConcept C41008148 @default.
- W4283523645 hasConcept C43214815 @default.
- W4283523645 hasConcept C44249647 @default.
- W4283523645 hasConcept C49937458 @default.
- W4283523645 hasConcept C53533937 @default.
- W4283523645 hasConcept C62520636 @default.
- W4283523645 hasConceptScore W4283523645C105795698 @default.
- W4283523645 hasConceptScore W4283523645C11413529 @default.
- W4283523645 hasConceptScore W4283523645C115961682 @default.
- W4283523645 hasConceptScore W4283523645C121332964 @default.
- W4283523645 hasConceptScore W4283523645C124101348 @default.
- W4283523645 hasConceptScore W4283523645C153180895 @default.
- W4283523645 hasConceptScore W4283523645C154945302 @default.
- W4283523645 hasConceptScore W4283523645C163258240 @default.
- W4283523645 hasConceptScore W4283523645C165838908 @default.
- W4283523645 hasConceptScore W4283523645C33923547 @default.
- W4283523645 hasConceptScore W4283523645C41008148 @default.
- W4283523645 hasConceptScore W4283523645C43214815 @default.
- W4283523645 hasConceptScore W4283523645C44249647 @default.
- W4283523645 hasConceptScore W4283523645C49937458 @default.
- W4283523645 hasConceptScore W4283523645C53533937 @default.
- W4283523645 hasConceptScore W4283523645C62520636 @default.
- W4283523645 hasLocation W42835236451 @default.
- W4283523645 hasOpenAccess W4283523645 @default.
- W4283523645 hasPrimaryLocation W42835236451 @default.
- W4283523645 hasRelatedWork W1498259939 @default.
- W4283523645 hasRelatedWork W1983610137 @default.
- W4283523645 hasRelatedWork W2097458023 @default.
- W4283523645 hasRelatedWork W2726222394 @default.
- W4283523645 hasRelatedWork W2750350980 @default.
- W4283523645 hasRelatedWork W2901148276 @default.
- W4283523645 hasRelatedWork W2990472155 @default.
- W4283523645 hasRelatedWork W3024593842 @default.
- W4283523645 hasRelatedWork W3140013985 @default.
- W4283523645 hasRelatedWork W2181817726 @default.
- W4283523645 hasVolume "607" @default.
- W4283523645 isParatext "false" @default.
- W4283523645 isRetracted "false" @default.
- W4283523645 workType "article" @default.