{"id":"W4307562325","doi":"10.1139/cjce-2022-0137","title":"Feature representation improved Faster R-CNN model for high-efficiency pavement crack detection","year":2022,"lang":"en","type":"article","venue":"Canadian Journal of Civil Engineering","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Feature (linguistics); Interpretability; Pattern recognition (psychology); Convolutional neural network; Artificial intelligence; Residual; Computer science; Channel (broadcasting); Representation (politics); Algorithm","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004626435,0.001387128,0.000899252,0.0008010974,0.0002447646,0.0006635727,0.001734546,0.0008442119,0.001914091],"category_scores_gemma":[0.001101579,0.0005246788,0.001089748,0.0006875356,0.0002848895,0.001560723,0.0005927597,0.001010976,0.0007277473],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001124514,"about_ca_system_score_gemma":0.001269066,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01982173,"about_ca_topic_score_gemma":0.01950764,"domain_scores_codex":[0.9997081,0.00002677834,0.0000133924,0.0001178679,0.00007805226,0.00005580267],"domain_scores_gemma":[0.9997018,0.00005400381,0.00004295773,0.00005313468,0.0001340264,0.00001405072],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001125342,0.00007811341,0.001861952,0.00006103353,0.0001055192,0.00009828837,0.00003131587,0.8131874,0.01266724,0.002527572,0.003588376,0.1656806],"study_design_scores_gemma":[0.000002319,0.000009946279,0.0001438441,0.000001565234,0.000008565974,0.00001126137,0.000001397953,0.9981654,0.001151886,0.0002870386,0.0002136078,0.000003086131],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05315029,0.0006378023,0.9388495,0.0002629164,0.0001116916,0.0000567735,0.0003227432,0.003899356,0.002708911],"genre_scores_gemma":[0.7950022,0.0005889589,0.1935608,0.0002758937,0.0000756815,0.0001761293,0.001357762,0.0003762248,0.008586313],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01982173,"threshold_uncertainty_score":0.03941274,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006508158085657644,"score_gpt":0.1879855660976601,"score_spread":0.1814774080120025,"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."}}