{"id":"W2997724457","doi":"10.1177/1475921719896813","title":"Multi-level feature fusion in densely connected deep-learning architecture and depth-first search for crack segmentation on images collected with smartphones","year":2020,"lang":"en","type":"article","venue":"Structural Health Monitoring","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Artificial intelligence; Segmentation; Feature (linguistics); Pooling; Convolutional neural network; Deep learning; Convolution (computer science); Pattern recognition (psychology); Layer (electronics); Set (abstract data type); Precision and recall; Data set; Key (lock); Transfer of learning; Computer vision; Artificial neural network","routes":{"ca_aff":true,"ca_fund":false,"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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001051298,0.0003532705,0.000372885,0.0001735997,0.0004619343,0.00008210503,0.0001149706,0.0001360343,0.000002611201],"category_scores_gemma":[0.00006422702,0.0002982186,0.00003724045,0.000440921,0.0000405217,0.0001412072,0.00003551182,0.0007658125,9.082649e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002841728,"about_ca_system_score_gemma":0.00004860499,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008980695,"about_ca_topic_score_gemma":0.0003018018,"domain_scores_codex":[0.9983182,0.00005248883,0.0002930523,0.0004148583,0.0002469232,0.0006745012],"domain_scores_gemma":[0.999289,0.000173287,0.00008053562,0.0001104601,0.0001103045,0.0002364452],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"observational","study_design_scores_codex":[0.001991633,0.00001592366,0.313255,0.003683481,0.0001401536,0.00006688894,0.03309105,0.3614402,0.08788216,0.0000193348,0.0001661435,0.198248],"study_design_scores_gemma":[0.003473299,0.0006856396,0.90568,0.0006411704,0.00001552004,0.00003552533,0.003036675,0.02672363,0.05902229,0.00001764238,0.0001403613,0.0005282573],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9920613,0.0005433288,0.004934409,0.0005451117,0.0006585906,0.0009418356,0.00002504038,0.0002777941,0.00001265675],"genre_scores_gemma":[0.9592833,0.0001407028,0.0397118,0.00006003543,0.0005980523,0.00006363355,0.00004124605,0.00007789931,0.0000233563],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.592425,"threshold_uncertainty_score":0.999947,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02641169714593535,"score_gpt":0.2802984647925388,"score_spread":0.2538867676466035,"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."}}