{"id":"W4413049812","doi":"10.1016/j.ins.2025.122561","title":"A context-aware multi-stream attentive convolutional neural network for surface defect segmentation","year":2025,"lang":"en","type":"article","venue":"Information Sciences","topic":"Industrial Vision Systems and Defect Detection","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia, Okanagan Campus","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada","keywords":"Convolutional neural network; Computer science; Context (archaeology); Segmentation; Artificial intelligence; Surface (topology); Artificial neural network; Pattern recognition (psychology); Machine learning; Geology; Mathematics","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":[0.0004725352,0.0000932303,0.0001099904,0.0001067806,0.0003506293,0.0001667769,0.00008852186,0.00007143611,0.00001398064],"category_scores_gemma":[0.0000435228,0.0000821565,0.00007867647,0.0004724772,0.00005775431,0.001079196,0.00001285418,0.00006366292,0.00003349371],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007728983,"about_ca_system_score_gemma":0.00004864407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006433534,"about_ca_topic_score_gemma":0.00003944716,"domain_scores_codex":[0.9992059,0.00002491651,0.0003246033,0.0000851719,0.0001768672,0.0001824907],"domain_scores_gemma":[0.999553,0.0001400092,0.00007855013,0.00005775385,0.0001430369,0.00002768463],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003642367,0.000008513537,0.004956266,0.00006780418,0.0000468597,1.032229e-7,0.0006412859,0.9340134,0.0004537457,0.001349917,0.01608669,0.04233901],"study_design_scores_gemma":[0.0009791323,0.00006844798,0.004534824,0.00005717378,0.00001327062,0.000002619295,0.001879977,0.9765788,0.001923666,0.0000840061,0.01373462,0.0001434281],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4753515,0.0001591805,0.5175972,0.00009541306,0.003449305,0.0009741376,0.00008331659,0.0002998595,0.001990074],"genre_scores_gemma":[0.9987848,0.000003044637,0.0008400831,0.0001304397,0.00006371616,0.00004746078,0.00003519244,0.000002508817,0.00009277494],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5234333,"threshold_uncertainty_score":0.3350245,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03628870791710499,"score_gpt":0.293265785071727,"score_spread":0.256977077154622,"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."}}