{"id":"W4413887382","doi":"10.1109/tgrs.2025.3604644","title":"River Ice Fine-Grained Segmentation: A GF-2 Satellite Image Dataset and Deep Learning Benchmark","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Geoscience and Remote Sensing","topic":"Arctic and Antarctic ice dynamics","field":"Earth and Planetary Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Key Research and Development Program of China; Natural Science Foundation of Ningbo; National Key Basic Research Program For Youth; National Natural Science Foundation of China","keywords":"Benchmark (surveying); GF(2); Satellite; Remote sensing; Deep learning; Computer science; Image segmentation; Artificial intelligence; Segmentation; Satellite image; Geology; Mathematics; Geodesy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002547284,0.0001471556,0.000135877,0.0001454702,0.0009467629,0.000125114,0.00007290982,0.0000563141,0.00007058165],"category_scores_gemma":[0.00001356724,0.0001258056,0.00003014249,0.0003749924,0.0004453042,0.000367986,0.00000205254,0.0002438627,0.00001321514],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008106653,"about_ca_system_score_gemma":0.00004054439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001757799,"about_ca_topic_score_gemma":0.001314798,"domain_scores_codex":[0.9989476,0.00006564458,0.0001719647,0.0003679798,0.0001734966,0.000273281],"domain_scores_gemma":[0.9994799,0.0002102205,0.00004949791,0.0001277706,0.0000324698,0.0001000941],"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":[0.00002946251,0.000005790798,0.0001941386,0.00002306699,0.00001149673,0.00001720583,0.0006576847,0.0004132336,0.0003157228,0.000002708908,0.00001402629,0.9983155],"study_design_scores_gemma":[0.0006454258,0.0001770406,0.02586803,0.0001800708,0.000114263,0.0001789091,0.002995529,0.963653,0.0005681882,0.000651614,0.004569272,0.0003986801],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2076229,0.0001552304,0.7894752,0.0008188094,0.0004546221,0.0001720548,0.00011963,0.00004815207,0.001133363],"genre_scores_gemma":[0.9064046,0.001437429,0.09020065,0.0007639314,0.00003159075,4.797664e-8,0.0001351042,0.000004018068,0.001022654],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9979168,"threshold_uncertainty_score":0.728183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007012240920840388,"score_gpt":0.2245506412209264,"score_spread":0.2175384003000861,"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."}}