{"id":"W4286560905","doi":"10.1038/s41598-022-16302-3","title":"End-to-end deep learning framework for printed circuit board manufacturing defect classification","year":2022,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":115,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Excellence Research Chairs, Government of Canada","keywords":"Rework; Bottleneck; Computer science; Convolutional neural network; Deep learning; Printed circuit board; Benchmark (surveying); Artificial intelligence; Reliability (semiconductor); Throughput; Object detection; Production line; Pattern recognition (psychology); Real-time computing; Reliability engineering; Embedded system; Engineering; Telecommunications","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002074529,0.0001623261,0.000208464,0.0003680331,0.0009857598,0.0003123477,0.0001259873,0.0001050503,0.0004298858],"category_scores_gemma":[0.0003003114,0.0001757169,0.0001980504,0.0005330441,0.00002534131,0.0001151615,0.00007315493,0.0004532549,0.00003771896],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003742765,"about_ca_system_score_gemma":0.00004180214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001884582,"about_ca_topic_score_gemma":0.000008140496,"domain_scores_codex":[0.9978142,0.00007880951,0.0005458263,0.0006209235,0.0005660999,0.0003740885],"domain_scores_gemma":[0.9989359,0.0001134994,0.0001877792,0.0005656679,0.00007215515,0.000125044],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00007024731,0.00007574352,0.001873434,0.0002114197,0.0001761709,0.0001666346,0.002607104,0.5507663,0.1741306,0.001359026,0.01153244,0.2570308],"study_design_scores_gemma":[0.0002197855,0.0001286234,0.001949107,0.00005176045,0.00004077498,0.0002294305,0.0009630854,0.01341931,0.08904137,0.005418879,0.8880363,0.0005015417],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7103263,0.00008616755,0.2601359,0.00003146132,0.02505439,0.001122511,0.000004595126,0.0007186842,0.002520022],"genre_scores_gemma":[0.9973345,5.706794e-7,0.0006022242,0.00001229373,0.0002424799,0.0003851031,0.00004184857,0.00004727536,0.001333686],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8765039,"threshold_uncertainty_score":0.7581767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03187696537167703,"score_gpt":0.2558360937193208,"score_spread":0.2239591283476437,"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."}}