{"id":"W2964603971","doi":"10.1109/lgrs.2019.2926530","title":"Filling Voids in Elevation Models Using a Shadow-Constrained Convolutional Neural Network","year":2019,"lang":"en","type":"article","venue":"IEEE Geoscience and Remote Sensing Letters","topic":"Landslides and related hazards","field":"Environmental Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"Fundamental Research Funds for the Central Universities; National Key Research and Development Program of China; Natural Science Foundation of Shandong Province","keywords":"Convolutional neural network; Shadow (psychology); Computer science; Digital elevation model; Artificial intelligence; Shadow mapping; Inference; Interpolation (computer graphics); Weighting; Elevation (ballistics); Inverse distance weighting; Computer vision; Key (lock); Pattern recognition (psychology); Image (mathematics); Bilinear interpolation; Mathematics; Multivariate interpolation; Remote sensing; Geometry; Geography","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004056422,0.0006184192,0.0004181395,0.0003715314,0.0002170308,0.0005423964,0.0009875933,0.00050779,0.001429641],"category_scores_gemma":[0.001477908,0.0004545323,0.0005323879,0.0004185469,0.0004754782,0.001320997,0.001139197,0.0008184505,0.0003174761],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007117565,"about_ca_system_score_gemma":0.0008592104,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01422302,"about_ca_topic_score_gemma":0.03060186,"domain_scores_codex":[0.9998497,0.00001896049,0.000007402041,0.00005049634,0.0000454717,0.0000279434],"domain_scores_gemma":[0.9996505,0.0001051896,0.00004736852,0.0001053853,0.00006791229,0.00002374493],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007338623,0.00004583879,0.002046874,0.00003671964,0.00004135573,0.00008929108,0.0000513657,0.8943887,0.007302523,0.002390606,0.00103596,0.09249735],"study_design_scores_gemma":[0.000001686831,0.000006407513,0.0002044557,0.000002324914,0.000003066861,0.000008909441,0.000003183143,0.9972733,0.001442198,0.0007823346,0.0002693336,0.000002935344],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1391964,0.0002932121,0.8540914,0.0003287345,0.00005539786,0.00003715514,0.0003814701,0.003242739,0.002373512],"genre_scores_gemma":[0.8674178,0.0001809047,0.1285603,0.0001402483,0.00003108901,0.0000298607,0.0006198492,0.0001740409,0.002845922],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01422302,"threshold_uncertainty_score":0.02828044,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01264248124707185,"score_gpt":0.2091243944320881,"score_spread":0.1964819131850162,"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."}}