{"id":"W3196201121","doi":"10.1155/2021/4428964","title":"A Novel Deep Convolutional Neural Network Based on ResNet‐18 and Transfer Learning for Detection of Wood Knot Defects","year":2021,"lang":"en","type":"article","venue":"Journal of Sensors","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":59,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Canadian Institute for Advanced Research","funders":"Natural Sciences and Engineering Research Council of Canada; Harbin Institute of Technology; China Postdoctoral Science Foundation","keywords":"Convolutional neural network; Artificial intelligence; Transfer of learning; Deep learning; Knot (papermaking); Pattern recognition (psychology); Computer science; Residual; Artificial neural network; Block (permutation group theory); Pooling; Feature extraction; Engineering; Algorithm; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000327128,0.0007822921,0.0004450565,0.0005675677,0.0001735279,0.0003211407,0.001118112,0.0005846984,0.001401308],"category_scores_gemma":[0.0003963836,0.000265272,0.0005686591,0.0002909831,0.0002964971,0.0009387965,0.0004664693,0.0006182138,0.0004098067],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005505054,"about_ca_system_score_gemma":0.0007565987,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005608701,"about_ca_topic_score_gemma":0.007245708,"domain_scores_codex":[0.9998318,0.00001465519,0.000006864108,0.0000468896,0.00006834384,0.0000314461],"domain_scores_gemma":[0.9998608,0.00002412851,0.00002203506,0.00002144929,0.00005775215,0.00001373515],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004194769,0.0004366972,0.004707586,0.0001744542,0.000233134,0.0004065419,0.00006185019,0.3139536,0.09962726,0.004690856,0.006118047,0.5691705],"study_design_scores_gemma":[0.000009769874,0.0001230461,0.0007899989,0.000005376192,0.00002348262,0.00009263593,0.000005220139,0.981306,0.01580665,0.0004953805,0.001328403,0.00001403055],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1339545,0.0009901777,0.8552975,0.0002660053,0.0002025229,0.0001193165,0.0002627092,0.004047875,0.00485938],"genre_scores_gemma":[0.8074076,0.0005875871,0.1809858,0.0002267877,0.00005119352,0.0001127061,0.0007331697,0.00009275458,0.009802361],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005608701,"threshold_uncertainty_score":0.01115215,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02363579184973762,"score_gpt":0.2287787644999611,"score_spread":0.2051429726502235,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}