{"id":"W4223586001","doi":"10.1016/j.jafr.2022.100308","title":"Weed detection in soybean crops using custom lightweight deep learning models","year":2022,"lang":"en","type":"article","venue":"Journal of Agriculture and Food Research","topic":"Smart Agriculture and AI","field":"Agricultural and Biological Sciences","cited_by":142,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"Ryerson University","keywords":"Weed; Deep learning; Convolutional neural network; Computer science; Artificial intelligence; Precision agriculture; Raspberry pi; Agriculture; Agricultural engineering; Machine learning; Pattern recognition (psychology); Embedded system; Agronomy; Engineering; Internet of Things","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001398909,0.000147161,0.0002690814,0.0001045534,0.000962677,0.0001318031,0.0002925172,0.0001175662,0.0000873],"category_scores_gemma":[0.00004583283,0.00004859272,0.0001329822,0.001365571,0.00004802841,0.0003447795,0.0002105527,0.001572014,0.000001789796],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001251649,"about_ca_system_score_gemma":0.00001649134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002345543,"about_ca_topic_score_gemma":0.001024014,"domain_scores_codex":[0.9975396,0.0005533549,0.0003878948,0.0002331429,0.0008586915,0.0004272581],"domain_scores_gemma":[0.9990922,0.0001906911,0.0001912804,0.00003220177,0.000332703,0.0001608816],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0001452476,0.0002849486,0.00246036,0.00001277455,0.00005674041,0.00007399481,0.00126028,0.002058047,0.9652753,0.0001748869,0.000895318,0.02730211],"study_design_scores_gemma":[0.004235778,0.02501458,0.399984,0.0003711939,0.0002033716,0.005550014,0.09853835,0.007793729,0.07772709,0.01108027,0.3672855,0.002216188],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9954262,0.00268139,0.00001026459,0.0008818128,0.0001473431,0.0002006865,0.000003781596,0.00001251374,0.0006360167],"genre_scores_gemma":[0.9987776,0.000114654,0.0000474361,0.0000403679,0.0007545483,0.00001221184,0.000006657067,0.000001550399,0.0002449383],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8875482,"threshold_uncertainty_score":0.740423,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06570843945550137,"score_gpt":0.2722391239592269,"score_spread":0.2065306845037256,"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."}}