{"id":"W4417169486","doi":"10.1109/jstars.2025.3642040","title":"SAM4CH4: Zero-Shot Methane Plume Mapping With Segment Anything and Vision-Language Models","year":2025,"lang":"","type":"article","venue":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","topic":"Atmospheric and Environmental Gas Dynamics","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Natural Resources Canada; Memorial University of Newfoundland","funders":"Natural Sciences and Engineering Research Council of Canada; European Space Agency","keywords":"Segmentation; Benchmark (surveying); Thresholding; Image segmentation; Plume; Scalability; Ground truth; Encoder; Deep learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007301811,0.00223018,0.001181335,0.001047103,0.000682416,0.001207986,0.004487136,0.001979118,0.005834869],"category_scores_gemma":[0.001875661,0.0007465901,0.001952508,0.0006649183,0.0009005844,0.002030059,0.002133156,0.002303785,0.002525209],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001415113,"about_ca_system_score_gemma":0.002151218,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02760564,"about_ca_topic_score_gemma":0.04270806,"domain_scores_codex":[0.9994294,0.0000719778,0.0000192125,0.0002508975,0.0001333273,0.00009519819],"domain_scores_gemma":[0.9996467,0.0001125761,0.00002633061,0.0001036759,0.00007231536,0.0000384415],"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":[0.0009677133,0.0004921355,0.00217966,0.0005874922,0.0003788373,0.0006253083,0.0002952074,0.2507735,0.04499923,0.01063631,0.07071258,0.617352],"study_design_scores_gemma":[0.00005625285,0.0001028977,0.0003247029,0.00001536956,0.00002915784,0.00008774987,0.00005156729,0.9707971,0.01421173,0.007446528,0.006837354,0.00003974518],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03042054,0.001196176,0.8380288,0.0006346477,0.0003249362,0.0003699571,0.004054808,0.120977,0.00399309],"genre_scores_gemma":[0.3129503,0.0006702152,0.6425946,0.001314611,0.0001539767,0.0006456056,0.02248936,0.00449907,0.01468235],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02760564,"threshold_uncertainty_score":0.05488992,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01491250687493407,"score_gpt":0.2326072785200382,"score_spread":0.2176947716451041,"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."}}