{"id":"W2900519402","doi":"10.1109/igarss.2018.8519171","title":"Toward the Use of Deep Learning for Topographic Feature Extraction from High Resolution Optical Satellite Imagery","year":2018,"lang":"en","type":"article","venue":"","topic":"Automated Road and Building Extraction","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Natural Resources Canada","funders":"Public Safety Canada","keywords":"Artificial intelligence; Computer science; Convolutional neural network; RGB color model; Deep learning; Feature extraction; Satellite imagery; Pixel; Remote sensing; Satellite; Confusion matrix; Feature (linguistics); Cohen's kappa; Pattern recognition (psychology); Artificial neural network; High resolution; Scale (ratio); Computer vision; Geology; Geography; Cartography; Machine learning; Engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001098064,0.0001062458,0.0001114846,0.00007204624,0.0001000579,0.00005429787,0.00006125971,0.0001542154,0.00004771958],"category_scores_gemma":[0.00007036752,0.00007639396,0.00007757612,0.0001581836,0.00006107066,0.0003149448,0.00001094452,0.0002178732,0.00001156795],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000286665,"about_ca_system_score_gemma":0.000003565294,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008261732,"about_ca_topic_score_gemma":0.00003383184,"domain_scores_codex":[0.9994023,0.00002276283,0.0001573431,0.0001363376,0.0001122639,0.0001689678],"domain_scores_gemma":[0.9994746,0.0002361875,0.00004591284,0.0001365933,0.00007750786,0.00002915811],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002795298,0.00007977666,0.001808748,0.0001230166,0.000269607,0.000004402424,0.0009158728,0.06470372,0.6224842,0.002264004,0.005729976,0.3013372],"study_design_scores_gemma":[0.0004423373,0.0001713961,0.1182047,0.00007667164,0.0001429825,0.00001758845,0.0002685294,0.685985,0.1273077,0.0005684808,0.0664766,0.0003379565],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5841224,0.0003549708,0.4132212,0.000298176,0.0008432241,0.0002001881,0.00000474764,0.0005933861,0.00036167],"genre_scores_gemma":[0.9729881,0.0002006267,0.02622334,0.00002167682,0.0003701319,0.00001111943,0.0000382017,0.00002330865,0.0001234915],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6212813,"threshold_uncertainty_score":0.3115255,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02495116974718405,"score_gpt":0.2417965853086747,"score_spread":0.2168454155614906,"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."}}