{"id":"W4225125287","doi":"10.1016/j.marpolbul.2022.113666","title":"A novel deep learning method for marine oil spill detection from satellite synthetic aperture radar imagery","year":2022,"lang":"en","type":"article","venue":"Marine Pollution Bulletin","topic":"Oil Spill Detection and Mitigation","field":"Environmental Science","cited_by":99,"is_retracted":false,"has_abstract":false,"ca_institutions":"Fisheries and Oceans Canada; Bedford Institute of Oceanography","funders":"","keywords":"Synthetic aperture radar; Remote sensing; Environmental science; Oil spill; Satellite; Deep learning; Computer science; Convolutional neural network; Artificial intelligence; Geology; Engineering; Environmental 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":[0.0005436728,0.0006977039,0.0006421431,0.0006756053,0.0002716991,0.0005143748,0.001057743,0.0008774277,0.001806291],"category_scores_gemma":[0.0009878122,0.0003835796,0.0006269782,0.0006480143,0.000259749,0.0006718682,0.001001362,0.001119917,0.0008184271],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004347276,"about_ca_system_score_gemma":0.001034599,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01007467,"about_ca_topic_score_gemma":0.01427013,"domain_scores_codex":[0.9998234,0.00002457178,0.00001041094,0.00004557165,0.00005851851,0.00003763473],"domain_scores_gemma":[0.9996834,0.00008121874,0.00002829525,0.00003709309,0.0001403069,0.00002985711],"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":[0.0001984403,0.0002114813,0.002088291,0.00009365034,0.0001322411,0.00008672721,0.00003450257,0.1823193,0.02304332,0.002360672,0.008405634,0.7810258],"study_design_scores_gemma":[0.00000443725,0.00001484486,0.0002401361,0.000003594882,0.000007638717,0.00001376798,0.000003052669,0.9968674,0.001862481,0.0004359634,0.0005429573,0.00000383118],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04599773,0.0009305151,0.9483076,0.0004536467,0.0002212616,0.00006289515,0.0003613819,0.00185099,0.001813874],"genre_scores_gemma":[0.464396,0.0009401168,0.5154301,0.0006615365,0.0002612536,0.0001414174,0.002039855,0.0002060236,0.01592371],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01007467,"threshold_uncertainty_score":0.02003205,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006056497023964068,"score_gpt":0.2059095307267809,"score_spread":0.1998530337028168,"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."}}