{"id":"W4393320716","doi":"10.1016/j.infrared.2024.105288","title":"Swin transformer network leveraging multi-dimensional features for defect depth prediction","year":2024,"lang":"en","type":"article","venue":"Infrared Physics & Technology","topic":"Thermography and Photoacoustic Techniques","field":"Engineering","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université Laval","funders":"National Natural Science Foundation of China","keywords":"Thermography; Computer science; Transformer; Planar; Artificial intelligence; Pattern recognition (psychology); Algorithm; Infrared; Optics; Voltage; Electrical engineering","routes":{"ca_aff":true,"ca_fund":false,"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":[0.00008882898,0.0002282788,0.0002023921,0.000185527,0.000101895,0.00003427617,0.0001620387,0.0003231348,0.00001036805],"category_scores_gemma":[0.000006344673,0.0002282655,0.000203571,0.0005987337,0.0000742761,0.0001243242,0.00001496681,0.0004440228,0.000003008709],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005161517,"about_ca_system_score_gemma":0.00002362123,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004398969,"about_ca_topic_score_gemma":0.000004874124,"domain_scores_codex":[0.999064,0.000009026026,0.000193497,0.0002617261,0.00009044215,0.000381258],"domain_scores_gemma":[0.9996013,0.0000915878,0.00001613527,0.0002210252,0.00003464495,0.00003532562],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00010501,0.0001456399,0.0008849613,0.001139887,0.001617402,0.00004463069,0.001408423,0.1332609,0.3033741,0.04083355,0.06465448,0.452531],"study_design_scores_gemma":[0.001080487,0.0003161415,0.001458406,0.0007412264,0.0003816234,0.00009696283,0.00009489671,0.2275691,0.2860857,0.4293914,0.05172179,0.001062285],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02826468,0.004963384,0.9554964,0.0000714922,0.001105066,0.0006853736,0.0004501041,0.008027671,0.0009358673],"genre_scores_gemma":[0.9855202,0.00005782673,0.01352093,0.00004834589,0.00029581,0.0003385258,0.00008022237,0.00007802433,0.00006007665],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9572555,"threshold_uncertainty_score":0.9308398,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01142174276773654,"score_gpt":0.2314482193550314,"score_spread":0.2200264765872949,"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."}}