{"id":"W4412105672","doi":"10.1016/j.compag.2025.110728","title":"Deep metric learning for soil organic matter prediction: A novel similarity-based approach using smartphone-captured images","year":2025,"lang":"en","type":"article","venue":"Computers and Electronics in Agriculture","topic":"Soil Geostatistics and Mapping","field":"Environmental Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ministry of Agriculture, Food and Rural Affairs; University of British Columbia; University of Guelph","funders":"Natural Sciences and Engineering Research Council of Canada; Ontario Agri-Food Innovation Alliance","keywords":"Similarity (geometry); Metric (unit); Artificial intelligence; Organic matter; Deep learning; Computer science; Pattern recognition (psychology); Machine learning; Computer vision; Engineering; Image (mathematics); Ecology; Biology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003290407,0.0005893576,0.0004431447,0.0006779365,0.0001541813,0.0004269087,0.0007445985,0.0005443383,0.0009697718],"category_scores_gemma":[0.00106679,0.0001765797,0.0004180622,0.0006539374,0.0001941174,0.0007929732,0.0006958625,0.0004941885,0.0004497262],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004037089,"about_ca_system_score_gemma":0.0004249561,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005655,"about_ca_topic_score_gemma":0.007403679,"domain_scores_codex":[0.9997961,0.00002978798,0.00001046794,0.00007224524,0.00006503705,0.00002629266],"domain_scores_gemma":[0.9997513,0.000052227,0.00003941102,0.000031129,0.0001060638,0.00001989685],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002330825,0.0003783498,0.009777359,0.0001405064,0.0001337758,0.0001857793,0.0001190351,0.2309812,0.04581026,0.00233231,0.003600622,0.7063078],"study_design_scores_gemma":[0.000004611806,0.00004148255,0.00139864,0.000003572333,0.00000832729,0.00002956529,0.00001438769,0.9929487,0.004244324,0.000801506,0.0004972276,0.000007802558],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1075052,0.0003621801,0.8880559,0.0001801422,0.0000488958,0.00007017505,0.000307002,0.001396774,0.00207368],"genre_scores_gemma":[0.79842,0.000256188,0.1971145,0.0001435026,0.00004371972,0.00008275423,0.0007349165,0.00006278804,0.003141736],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005655,"threshold_uncertainty_score":0.01124418,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006887570174839554,"score_gpt":0.2039204190930687,"score_spread":0.1970328489182291,"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."}}