{"id":"W3212006298","doi":"10.1109/jstars.2021.3127728","title":"Identifying Building Rooftops in Hyperspectral Imagery Using CNN With Pure Pixel Index","year":2021,"lang":"en","type":"article","venue":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","topic":"Remote-Sensing Image Classification","field":"Engineering","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Major International Joint Research Programme; Chinese Academy of Sciences; National Natural Science Foundation of China","keywords":"Hyperspectral imaging; Artificial intelligence; Computer science; Convolutional neural network; Pattern recognition (psychology); Support vector machine; Pixel; Feature extraction; Deep learning; Classifier (UML); Remote sensing; Kernel (algebra); Computer vision; Mathematics; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003000992,0.0001917609,0.000341741,0.0004201044,0.0001016056,0.0001677159,0.00006779047,0.0001427188,0.000001329094],"category_scores_gemma":[0.00009542443,0.0001961096,0.00003807108,0.00136018,0.00005014644,0.000249063,0.00001445939,0.0007514493,2.406427e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000214681,"about_ca_system_score_gemma":0.0002038687,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002521068,"about_ca_topic_score_gemma":0.0002809363,"domain_scores_codex":[0.9985298,0.00004867346,0.0006098077,0.000205663,0.000294974,0.0003111103],"domain_scores_gemma":[0.9990836,0.00007789159,0.0001818469,0.0001744283,0.0004110737,0.00007113394],"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":[0.000018301,0.00001522422,0.00107376,0.00007022048,0.00003897185,0.0001934422,0.0003796893,0.08509642,0.8720913,0.00008102592,0.000006129731,0.04093551],"study_design_scores_gemma":[0.001120444,0.00001860549,0.07189916,0.0008200471,0.00004880912,0.001112617,0.0006327175,0.8411,0.08151425,0.001181273,0.0001752017,0.0003768916],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.789244,0.0002602547,0.209812,0.0001443165,0.0001943964,0.00008619808,6.374516e-7,0.00003248945,0.0002256511],"genre_scores_gemma":[0.6887756,0.0001353229,0.3108175,0.00002879164,0.0001943628,2.462203e-8,0.000002012939,0.00003006076,0.00001627461],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7905771,"threshold_uncertainty_score":0.7997118,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03037394208202287,"score_gpt":0.241177609937393,"score_spread":0.2108036678553701,"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."}}