{"id":"W3087478532","doi":"10.30897/ijegeo.737993","title":"Comparison of Fully Convolutional Networks (FCN) and U-Net for Road Segmentation from High Resolution Imageries","year":2020,"lang":"en","type":"article","venue":"International Journal of Environment and Geoinformatics","topic":"Automated Road and Building Extraction","field":"Engineering","cited_by":65,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Toronto","keywords":"Computer science; Artificial intelligence; Pixel; Segmentation; Convolutional neural network; Image segmentation; Computer vision; Shadow (psychology); Context (archaeology); Pattern recognition (psychology); Set (abstract data type); Deep learning; Image resolution; Geography","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001203549,0.001387515,0.0006866065,0.001279981,0.0004249247,0.0008264076,0.001228842,0.00173942,0.001343637],"category_scores_gemma":[0.002127616,0.0003584994,0.0005984985,0.0009392094,0.0003137109,0.001597465,0.0004865864,0.0007749862,0.00044163],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001369018,"about_ca_system_score_gemma":0.001204484,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04370835,"about_ca_topic_score_gemma":0.04252049,"domain_scores_codex":[0.9995807,0.00005923279,0.00002742967,0.0001130571,0.00009705281,0.0001226264],"domain_scores_gemma":[0.9994118,0.0002117038,0.00004018679,0.00005184454,0.0002290258,0.00005554967],"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":[0.002386133,0.0005485026,0.007742109,0.0006477108,0.0005096819,0.0003279757,0.0001299728,0.4060181,0.01168794,0.002021823,0.01239873,0.5555813],"study_design_scores_gemma":[0.00002100363,0.0001736722,0.003135527,0.00003543167,0.00006158843,0.00003900017,0.00005411431,0.9896036,0.005499787,0.0004700748,0.0008890625,0.00001719889],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7957518,0.01641509,0.1529106,0.001321975,0.0009714132,0.0002542918,0.001993736,0.009238622,0.02114253],"genre_scores_gemma":[0.9220901,0.002622634,0.06532643,0.0003149269,0.00008353889,0.00008692654,0.00323127,0.0001942462,0.006049982],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04370835,"threshold_uncertainty_score":0.08690786,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01122545326914831,"score_gpt":0.2358083245390289,"score_spread":0.2245828712698806,"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."}}