{"id":"W2149617283","doi":"10.1109/tip.2004.833101","title":"Iterative Multimodel Subimage Binarization for Handwritten Character Segmentation","year":2004,"lang":"en","type":"article","venue":"IEEE Transactions on Image Processing","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Artificial intelligence; Computer science; Segmentation; Image segmentation; Pattern recognition (psychology); Character (mathematics); Computer vision; Image (mathematics); Image processing; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.000195959,0.000228805,0.0001798296,0.0003346586,0.0005502722,0.0006876565,0.0002904649,0.0000967443,0.00001299503],"category_scores_gemma":[0.00001015857,0.0002307634,0.0001068299,0.0004530155,0.00006480428,0.003470985,0.000002639157,0.0001917627,0.00004008236],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001716498,"about_ca_system_score_gemma":0.0001265702,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001036059,"about_ca_topic_score_gemma":0.000005797276,"domain_scores_codex":[0.9985727,0.00003990808,0.000334408,0.0005093731,0.0002485387,0.0002950725],"domain_scores_gemma":[0.9990253,0.00004062126,0.0001508439,0.0002353865,0.0004644401,0.00008338597],"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":[0.00005874885,0.0005018765,0.000002115756,0.0001600621,0.00002560973,0.000007306813,0.004341022,0.001274824,0.3736484,0.0001696892,0.0000231583,0.6197872],"study_design_scores_gemma":[0.00128514,0.0001806926,0.00002082474,0.0002071226,0.00002354596,0.00001843418,0.00006990363,0.1090678,0.885268,0.003511936,0.00004356951,0.000303093],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003293558,0.00002223936,0.9940574,0.000902108,0.0001755733,0.0007223556,0.00002888256,0.0006806658,0.0001172582],"genre_scores_gemma":[0.573695,0.00001136668,0.4253237,0.0004138678,0.00003491822,0.0003900653,0.00001342219,0.00002356908,0.00009410523],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6194841,"threshold_uncertainty_score":0.9410259,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01888049150970877,"score_gpt":0.2821144217659209,"score_spread":0.2632339302562121,"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."}}