Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366264295> ?p ?o ?g. }
- W4366264295 endingPage "1107" @default.
- W4366264295 startingPage "1107" @default.
- W4366264295 abstract "Surface defect identification plays a vital role in defective component rapid screening tasks in optics-related industries. However, the weakness and complexity of optical surface defects pose considerable challenges to their effective identification. To this end, a deep network based on multi-scale mixed kernels and structural re-parameterization is proposed to identify four manufacturing and two non-manufacturing optical surface defects. First, we design a multi-size mixed convolutional kernel with multiple receptive fields to extract rich shallow features for characterizing the defects with varying scales and irregular shapes. Then, we design an asymmetric mixed kernel integrating square, horizontal, vertical, and point convolutions to capture rotationally robust middle-and-deep features. Moreover, a structural re-parameterization strategy is introduced to equivalently convert the multi-branch architecture in the training phase into a deploy-friendly single-branch architecture in the inference phase, so that the model can obtain higher inference speed without losing any performance. Experiments on an optical surface defect dataset demonstrate that the proposed method is efficient and effective. It achieves a remarkable accuracy of 97.39% and an ultra-fast inference speed of 201.76 frames/second with only 5.23M parameters. Such a favorable accuracy-speed trade-off is capable of meeting the requirements of real-world optical surface defect identification applications." @default.
- W4366264295 created "2023-04-20" @default.
- W4366264295 creator A5028883593 @default.
- W4366264295 creator A5041572932 @default.
- W4366264295 creator A5052169513 @default.
- W4366264295 creator A5054005757 @default.
- W4366264295 creator A5056842713 @default.
- W4366264295 creator A5068252474 @default.
- W4366264295 date "2023-05-18" @default.
- W4366264295 modified "2023-10-18" @default.
- W4366264295 title "Efficient surface defect identification for optical components via multi-scale mixed Kernels and structural re-parameterization" @default.
- W4366264295 cites W2018930433 @default.
- W4366264295 cites W2553890187 @default.
- W4366264295 cites W2739024792 @default.
- W4366264295 cites W2761796288 @default.
- W4366264295 cites W2804878141 @default.
- W4366264295 cites W2920311927 @default.
- W4366264295 cites W2960766592 @default.
- W4366264295 cites W2972841284 @default.
- W4366264295 cites W3011369920 @default.
- W4366264295 cites W3017134824 @default.
- W4366264295 cites W3130840090 @default.
- W4366264295 cites W3160562437 @default.
- W4366264295 cites W3162778129 @default.
- W4366264295 cites W3213693699 @default.
- W4366264295 cites W4206041790 @default.
- W4366264295 cites W4210846093 @default.
- W4366264295 cites W4283273427 @default.
- W4366264295 cites W4304754988 @default.
- W4366264295 doi "https://doi.org/10.1364/josaa.484936" @default.
- W4366264295 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37706764" @default.
- W4366264295 hasPublicationYear "2023" @default.
- W4366264295 type Work @default.
- W4366264295 citedByCount "0" @default.
- W4366264295 crossrefType "journal-article" @default.
- W4366264295 hasAuthorship W4366264295A5028883593 @default.
- W4366264295 hasAuthorship W4366264295A5041572932 @default.
- W4366264295 hasAuthorship W4366264295A5052169513 @default.
- W4366264295 hasAuthorship W4366264295A5054005757 @default.
- W4366264295 hasAuthorship W4366264295A5056842713 @default.
- W4366264295 hasAuthorship W4366264295A5068252474 @default.
- W4366264295 hasConcept C11413529 @default.
- W4366264295 hasConcept C114614502 @default.
- W4366264295 hasConcept C116834253 @default.
- W4366264295 hasConcept C121332964 @default.
- W4366264295 hasConcept C154945302 @default.
- W4366264295 hasConcept C168167062 @default.
- W4366264295 hasConcept C2524010 @default.
- W4366264295 hasConcept C2776214188 @default.
- W4366264295 hasConcept C2776799497 @default.
- W4366264295 hasConcept C33923547 @default.
- W4366264295 hasConcept C41008148 @default.
- W4366264295 hasConcept C44280652 @default.
- W4366264295 hasConcept C45347329 @default.
- W4366264295 hasConcept C50644808 @default.
- W4366264295 hasConcept C59822182 @default.
- W4366264295 hasConcept C62520636 @default.
- W4366264295 hasConcept C74193536 @default.
- W4366264295 hasConcept C81363708 @default.
- W4366264295 hasConcept C86803240 @default.
- W4366264295 hasConcept C97355855 @default.
- W4366264295 hasConceptScore W4366264295C11413529 @default.
- W4366264295 hasConceptScore W4366264295C114614502 @default.
- W4366264295 hasConceptScore W4366264295C116834253 @default.
- W4366264295 hasConceptScore W4366264295C121332964 @default.
- W4366264295 hasConceptScore W4366264295C154945302 @default.
- W4366264295 hasConceptScore W4366264295C168167062 @default.
- W4366264295 hasConceptScore W4366264295C2524010 @default.
- W4366264295 hasConceptScore W4366264295C2776214188 @default.
- W4366264295 hasConceptScore W4366264295C2776799497 @default.
- W4366264295 hasConceptScore W4366264295C33923547 @default.
- W4366264295 hasConceptScore W4366264295C41008148 @default.
- W4366264295 hasConceptScore W4366264295C44280652 @default.
- W4366264295 hasConceptScore W4366264295C45347329 @default.
- W4366264295 hasConceptScore W4366264295C50644808 @default.
- W4366264295 hasConceptScore W4366264295C59822182 @default.
- W4366264295 hasConceptScore W4366264295C62520636 @default.
- W4366264295 hasConceptScore W4366264295C74193536 @default.
- W4366264295 hasConceptScore W4366264295C81363708 @default.
- W4366264295 hasConceptScore W4366264295C86803240 @default.
- W4366264295 hasConceptScore W4366264295C97355855 @default.
- W4366264295 hasFunder F4320321001 @default.
- W4366264295 hasFunder F4320322163 @default.
- W4366264295 hasIssue "6" @default.
- W4366264295 hasLocation W43662642951 @default.
- W4366264295 hasLocation W43662642952 @default.
- W4366264295 hasOpenAccess W4366264295 @default.
- W4366264295 hasPrimaryLocation W43662642951 @default.
- W4366264295 hasRelatedWork W2550196753 @default.
- W4366264295 hasRelatedWork W2766634277 @default.
- W4366264295 hasRelatedWork W2949189996 @default.
- W4366264295 hasRelatedWork W2959500052 @default.
- W4366264295 hasRelatedWork W2971742398 @default.
- W4366264295 hasRelatedWork W3006085271 @default.
- W4366264295 hasRelatedWork W3085075731 @default.
- W4366264295 hasRelatedWork W3200060857 @default.
- W4366264295 hasRelatedWork W4312417841 @default.
- W4366264295 hasRelatedWork W4323831447 @default.