{"id":"W3121754326","doi":"10.1109/jstars.2021.3053603","title":"Reconstruction Bias U-Net for Road Extraction From Optical Remote Sensing Images","year":2021,"lang":"en","type":"article","venue":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","topic":"Automated Road and Building Extraction","field":"Engineering","cited_by":106,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Science Foundation of Fujian Province; National Natural Science Foundation of China","keywords":"Computer science; Convolutional neural network; Artificial intelligence; Decoding methods; Segmentation; Remote sensing; Information extraction; Feature extraction; Deep learning; Image segmentation; Computer vision; Intelligent transportation system; Pattern recognition (psychology); Geography; Telecommunications","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000248756,0.0001560951,0.000280914,0.0001835339,0.0001402381,0.0001099199,0.0000325711,0.0002054799,0.000003036404],"category_scores_gemma":[0.0001596405,0.0001612279,0.00005836451,0.0004319881,0.0000299619,0.0001950979,0.000006647341,0.0004830284,5.465017e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009195687,"about_ca_system_score_gemma":0.00008872935,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005845193,"about_ca_topic_score_gemma":0.0001241581,"domain_scores_codex":[0.9988313,0.00003312163,0.0005739568,0.0001713963,0.0001795571,0.0002106612],"domain_scores_gemma":[0.999062,0.0001611691,0.0001930947,0.0001112535,0.0004169712,0.00005552384],"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":[0.00001684715,0.000004765367,0.0000108194,0.00002420275,0.00003894217,0.00001863132,0.00006334616,0.008974711,0.3635036,0.000009634343,0.00005122296,0.6272833],"study_design_scores_gemma":[0.0007543085,0.00002541409,0.01187427,0.0003488195,0.00008013521,0.0008393519,0.0001792949,0.7642205,0.2168905,0.002150206,0.00239879,0.0002384689],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.6751273,0.0002107874,0.3231004,0.0002075933,0.0009517015,0.00009193959,0.000004475038,0.00006873915,0.000237032],"genre_scores_gemma":[0.4819128,0.0005334623,0.5167807,0.00003511614,0.0006538158,2.036495e-8,0.00001638307,0.00002789062,0.00003980625],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7552457,"threshold_uncertainty_score":0.6574682,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0286536655783256,"score_gpt":0.2456760576844232,"score_spread":0.2170223921060976,"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."}}