{"id":"W4398786207","doi":"10.1002/2688-8319.12324","title":"Detecting flowers on imagery with computer vision to improve continental scale grassland biodiversity surveying","year":2024,"lang":"en","type":"article","venue":"Ecological Solutions and Evidence","topic":"Plant and animal studies","field":"Agricultural and Biological Sciences","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Instituto de Física de Cantabria; European Commission; Institute for Catastrophic Loss Reduction","keywords":"Biodiversity; Scale (ratio); Vegetation (pathology); Grassland; Computer science; Abundance (ecology); Precision and recall; Remote sensing; Geography; Cartography; Artificial intelligence; Ecology; Biology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002897423,0.0001038929,0.000124381,0.00001504598,0.0005854277,0.0001344482,0.00008389709,0.00004125258,0.00005074276],"category_scores_gemma":[0.00005417296,0.00003633563,0.00003882838,0.0002499034,0.00005653704,0.0001179449,0.0001915706,0.0000985055,0.00009823309],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002802578,"about_ca_system_score_gemma":0.000002679024,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002149293,"about_ca_topic_score_gemma":0.002511794,"domain_scores_codex":[0.9991596,0.00004117621,0.00008882045,0.0003462512,0.0001056569,0.0002584318],"domain_scores_gemma":[0.9993749,0.0004466771,0.00001443055,0.00001483347,0.00002668979,0.0001224415],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0008057017,0.0003375357,0.4722942,0.00005485598,0.00006803621,0.0002843619,0.0004603428,0.00005342476,0.3046948,0.00006949815,0.01313455,0.2077426],"study_design_scores_gemma":[0.00003339058,0.001909999,0.9962592,0.000158552,0.000008482742,0.000009077966,0.0001207236,0.0003376497,0.0001873279,0.00001244374,0.0008297542,0.000133442],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9969184,0.0001741131,0.0001026004,0.002337602,0.0001041147,0.0001652928,0.00003807186,0.00008423748,0.00007555633],"genre_scores_gemma":[0.9991844,0.00004458565,0.0003649403,0.0002681962,0.00009567937,0.000008611738,0.000002018613,3.305565e-7,0.00003123503],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5239649,"threshold_uncertainty_score":0.4502696,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04815915229442592,"score_gpt":0.230322414639387,"score_spread":0.1821632623449611,"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."}}