{"id":"W4285253164","doi":"10.1109/jstars.2022.3177579","title":"Wet-GC: A Novel Multimodel Graph Convolutional Approach for Wetland Classification Using Sentinel-1 and 2 Imagery With Limited Training Samples","year":2022,"lang":"en","type":"article","venue":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","topic":"Land Use and Ecosystem Services","field":"Environmental Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"Natural Sciences and Engineering Research Council of Canada; Department of Environment and Conservation, Government of Newfoundland and Labrador; European Space Agency","keywords":"Computer science; Wetland; Remote sensing; Convolutional neural network; Synthetic aperture radar; Bottleneck; Artificial intelligence; Graph; Deep learning; Contextual image classification; Multispectral image; Image resolution; Pattern recognition (psychology); Environmental science; Ecology; Image (mathematics); Geology","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":[0.0003659832,0.001254772,0.0006205234,0.0009973082,0.0003178147,0.0005344368,0.001858965,0.0008818355,0.001679619],"category_scores_gemma":[0.0006859011,0.0003450854,0.0007061216,0.0008346062,0.0003634614,0.001168803,0.0008914631,0.0009141092,0.000515316],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001006135,"about_ca_system_score_gemma":0.001273989,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03106576,"about_ca_topic_score_gemma":0.05396738,"domain_scores_codex":[0.9998223,0.00001799435,0.00000716112,0.00005815149,0.00004677642,0.00004766004],"domain_scores_gemma":[0.9998316,0.00004093895,0.00002407806,0.00003113898,0.0000499619,0.00002227685],"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":[0.0002202874,0.0001985751,0.004061785,0.0001373386,0.0002233079,0.0002263324,0.00008736125,0.3575782,0.01927234,0.00539282,0.0113203,0.6012813],"study_design_scores_gemma":[0.000005847212,0.00002022064,0.0003382618,0.000003890698,0.00001223892,0.00002462473,0.000007511979,0.9961515,0.001450759,0.001203308,0.0007765455,0.000005230421],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07959722,0.001032496,0.9075706,0.000474974,0.0001389381,0.0001664635,0.0006783614,0.006503444,0.003837501],"genre_scores_gemma":[0.6172304,0.0006835404,0.3645587,0.0007342394,0.00009642227,0.0001813175,0.003913866,0.0003536303,0.01224788],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03106576,"threshold_uncertainty_score":0.0617699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0711564252892243,"score_gpt":0.235333452496089,"score_spread":0.1641770272068647,"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."}}