{"id":"W2127184823","doi":"10.1109/euvip.2010.5699135","title":"Text extraction from degraded document images","year":2010,"lang":"en","type":"article","venue":"","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Skeletonization; Computer science; Cluster analysis; Artificial intelligence; Segmentation; Image segmentation; Pattern recognition (psychology); Pixel; Coherence (philosophical gambling strategy); Historical document; Image (mathematics); Computer vision; 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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001541706,0.0001018473,0.00008922747,0.00008476647,0.00007375456,0.0002759032,0.0004882818,0.00007731825,0.001234467],"category_scores_gemma":[0.00003364866,0.00008657813,0.00005031329,0.0001414193,0.00002981686,0.0009147764,0.0001093747,0.0002280204,0.0006772273],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000015397,"about_ca_system_score_gemma":0.00002397618,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003032588,"about_ca_topic_score_gemma":0.00007221917,"domain_scores_codex":[0.9991455,0.00002855788,0.0001628335,0.0003041944,0.0001956233,0.0001632679],"domain_scores_gemma":[0.9992416,0.00009236814,0.00005591556,0.0004575438,0.00006841477,0.00008417964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000001476845,0.00005353712,0.0002039845,0.000001379678,0.000007640107,0.000008559381,0.00006946888,3.050683e-8,0.2957268,0.01478316,0.0109108,0.6782331],"study_design_scores_gemma":[0.0001339153,0.00002599993,0.00475003,0.000004841229,0.000003093226,0.00001544737,0.00001115826,0.0003833133,0.9211069,0.05754509,0.01586462,0.0001555642],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03820861,0.00002286919,0.930302,0.001884608,0.0005246959,0.0001484729,0.000002485852,0.001248553,0.02765777],"genre_scores_gemma":[0.6159372,0.00001138986,0.3824435,0.000341129,0.00008231762,0.00002813365,0.00000429436,0.000005439946,0.001146515],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6780776,"threshold_uncertainty_score":0.9996786,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01075154259886924,"score_gpt":0.2694868028965848,"score_spread":0.2587352602977156,"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."}}