{"id":"W4399424031","doi":"10.1117/12.3013684","title":"A few-shot learning approach for the segmentation of subsurface defects in thermography images of concrete structures","year":2024,"lang":"en","type":"article","venue":"","topic":"Thermography and Photoacoustic Techniques","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Université Laval","funders":"","keywords":"Thermography; Shot (pellet); Segmentation; Computer science; Image segmentation; Artificial intelligence; Computer vision; Single shot; Pattern recognition (psychology); Materials science; Optics; Infrared; Physics","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":[0.0008733316,0.0007324063,0.0007150972,0.001349806,0.0003519165,0.0006431383,0.001318496,0.001173784,0.0007612925],"category_scores_gemma":[0.001728648,0.0003623002,0.0007632112,0.0006554798,0.0006097339,0.000968546,0.0007716807,0.001091375,0.000441658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005776869,"about_ca_system_score_gemma":0.000846855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00358339,"about_ca_topic_score_gemma":0.007953458,"domain_scores_codex":[0.9996055,0.00005671308,0.00001969945,0.0001877648,0.00008244986,0.00004795902],"domain_scores_gemma":[0.999401,0.0001904032,0.00008388666,0.00009428969,0.0001758701,0.00005455788],"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":[0.0003988146,0.0004577143,0.003802664,0.000247538,0.0001417522,0.0002712307,0.0003059521,0.1909805,0.0876761,0.00289787,0.002621368,0.7101985],"study_design_scores_gemma":[0.000004014378,0.00008713496,0.001172794,0.000007368265,0.00001479989,0.0001024611,0.00003517443,0.9875739,0.009523084,0.001058308,0.0004110608,0.000009778191],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.107761,0.0004140612,0.8890094,0.0001595744,0.00005403917,0.0001294004,0.0001514654,0.001307547,0.001013546],"genre_scores_gemma":[0.5838403,0.0002958806,0.4116963,0.0001649347,0.00006021808,0.0001208517,0.0008046785,0.0001127151,0.002904107],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00358339,"threshold_uncertainty_score":0.00712508,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01182069369094115,"score_gpt":0.2396414001344117,"score_spread":0.2278207064434705,"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."}}