{"id":"W3033645921","doi":"10.1016/j.autcon.2020.103291","title":"Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning","year":2020,"lang":"en","type":"article","venue":"Automation in Construction","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":395,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Manitoba","funders":"Natural Sciences and Engineering Research Council of Canada; Research Manitoba","keywords":"Pixel; Intersection (aeronautics); Segmentation; Convolutional neural network; Artificial intelligence; Computer science; Algorithm; Bounding overwatch; Computer vision; Pattern recognition (psychology); Engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003846747,0.0009109309,0.0008764291,0.001456997,0.0003121048,0.001093331,0.001269622,0.001320322,0.002016007],"category_scores_gemma":[0.0007635331,0.0005530608,0.0007889272,0.001000255,0.0004551352,0.00133556,0.001525741,0.001084278,0.0009929962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005779281,"about_ca_system_score_gemma":0.0009366451,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007709621,"about_ca_topic_score_gemma":0.0174143,"domain_scores_codex":[0.9997118,0.00002067613,0.000009865076,0.0001128353,0.00007817084,0.00006661967],"domain_scores_gemma":[0.9996518,0.00009266186,0.00004342154,0.00006462723,0.0001099193,0.00003771133],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004267711,0.0003991559,0.005867916,0.0001788747,0.0001713886,0.0001879059,0.0001452549,0.3149495,0.08658944,0.003994914,0.004660883,0.582428],"study_design_scores_gemma":[0.000003577741,0.00001640546,0.001043377,0.000007117163,0.00001149213,0.00002602767,0.00001201402,0.9912971,0.00568962,0.001436504,0.0004499855,0.000006858774],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08886816,0.0003451424,0.9048719,0.0001964268,0.00004082156,0.00004712939,0.0004151356,0.002790439,0.002424737],"genre_scores_gemma":[0.6427534,0.0003439306,0.3480432,0.0002599967,0.00006141904,0.00006728626,0.001738829,0.0003527667,0.006379126],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007709621,"threshold_uncertainty_score":0.01532948,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04331559413787201,"score_gpt":0.2807973856158583,"score_spread":0.2374817914779863,"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."}}