{"id":"W4402330136","doi":"10.1016/j.jas.2024.106053","title":"Deep learning-based detection of qanat underground water distribution systems using HEXAGON spy satellite imagery","year":2024,"lang":"en","type":"article","venue":"Journal of Archaeological Science","topic":"Water management and technologies","field":"Agricultural and Biological Sciences","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Amgen (Canada)","funders":"HORIZON EUROPE Marie Sklodowska-Curie Actions; U.S. Geological Survey","keywords":"Satellite imagery; Satellite; Remote sensing; Deep learning; Distribution (mathematics); Geology; Geography; Cartography; Computer science; Artificial intelligence; Engineering; Mathematics","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":[0.0001422304,0.0005577474,0.0002412229,0.000858312,0.0001409797,0.000382676,0.0004643007,0.0003138707,0.0008769872],"category_scores_gemma":[0.0003309705,0.0002065794,0.0004152016,0.0005373642,0.0002185327,0.0003543578,0.0003752599,0.000334473,0.0003521839],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004119604,"about_ca_system_score_gemma":0.0004178988,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01788413,"about_ca_topic_score_gemma":0.02331194,"domain_scores_codex":[0.9998966,0.000008451891,0.00000433958,0.00003182394,0.00002067662,0.00003816062],"domain_scores_gemma":[0.9999002,0.00001880459,0.00001643717,0.00001327555,0.0000381258,0.00001309822],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0006541056,0.0003677902,0.05855843,0.0002108153,0.0002584021,0.0008425009,0.0002663096,0.4081411,0.07635444,0.0009199706,0.01047029,0.4429559],"study_design_scores_gemma":[0.0000148545,0.00005584861,0.01913266,0.00001196702,0.00002164467,0.00006800202,0.0001005717,0.9660642,0.01319063,0.0002578995,0.001072056,0.000009663943],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9371364,0.0003066475,0.0550852,0.0002065296,0.00008019681,0.00008172062,0.001220621,0.002145485,0.003737228],"genre_scores_gemma":[0.9660799,0.0001086207,0.0289633,0.00005490248,0.00001621887,0.0000197927,0.002469562,0.00003895438,0.002248733],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01788413,"threshold_uncertainty_score":0.03556007,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02093653154018369,"score_gpt":0.2299417410526789,"score_spread":0.2090052095124953,"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."}}