{"id":"W4378191359","doi":"10.1016/j.ndteint.2023.102884","title":"PHM-IRNET: Self-training thermal segmentation approach for thermographic inspection of industrial components","year":2023,"lang":"en","type":"article","venue":"NDT & E International","topic":"Thermography and Photoacoustic Techniques","field":"Engineering","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université Laval","funders":"","keywords":"Segmentation; Artificial intelligence; Process (computing); Computer science; Artificial neural network; Task (project management); Piping; Ground truth; Pattern recognition (psychology); Deep learning; Nondestructive testing; Computer vision; Engineering; Mechanical 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005184964,0.001753224,0.001279721,0.001425781,0.0004725102,0.0005999772,0.002277362,0.001872546,0.004211166],"category_scores_gemma":[0.0007814249,0.0007754416,0.001235257,0.001072511,0.0003411215,0.0009830652,0.0009294527,0.001287729,0.002695767],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000676615,"about_ca_system_score_gemma":0.001045822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01073379,"about_ca_topic_score_gemma":0.02252794,"domain_scores_codex":[0.9996709,0.0000337251,0.00001290525,0.0001541028,0.00007281725,0.00005568629],"domain_scores_gemma":[0.9997863,0.00006385841,0.00002229416,0.00004504009,0.0000634106,0.00001906446],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005263136,0.0004267733,0.001715853,0.0002839699,0.0001833374,0.0001702506,0.0001024554,0.1330938,0.05463118,0.001175814,0.02461247,0.7830777],"study_design_scores_gemma":[0.00001407155,0.00007432211,0.0007841857,0.00001548329,0.00003337953,0.00007579188,0.00001937485,0.9808987,0.01432235,0.0008418214,0.002906001,0.0000146129],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04105684,0.001164861,0.9125057,0.0001744544,0.0002214969,0.0001691379,0.001440151,0.04002326,0.0032442],"genre_scores_gemma":[0.3021751,0.0005932982,0.6753684,0.0005811462,0.0001360519,0.00039393,0.005877326,0.00182426,0.01305053],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01073379,"threshold_uncertainty_score":0.02134264,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05161564790717314,"score_gpt":0.2545552835374891,"score_spread":0.202939635630316,"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."}}