@InProceedings{OliveiraJúniorKapFreCarSab:2004:HaReMu,
               author = "Oliveira J{\'u}nior, Jos{\'e} Josemar and Kapp, Marcelo 
                         Nepomoceno and Freitas, Cinthia Obladen de Almendra and Carvalho, 
                         Jo{\~a}o Marques de and Sabourin, Robert",
          affiliation = "Universidade Federal de Campina Grande, Coordena{\c{c}}{\~a}o de 
                         P{\'o}s-Gradua{\c{c}}{\~a}o em Engenharia El{\'e}trica, Caixa 
                         Postal 10105, 58109-970, Campina Grande, PB - Brazil and 
                         Pont{\'{\i}}ficia Universidade Cat{\'o}lica do Paran{\'a}, R. 
                         Imaculada Concei{\c{c}}{\~a}o 1155, 80215-901, Curitiba, PR - 
                         Brazil and {\`E}cole de Technologie Superieure, 1100 Rue Notre 
                         Dame Ouest, H3C 1K3, Montreal, QC - Canada",
                title = "Handwritten Recognition with Multiple Classifiers for Restricted 
                         Lexicon",
            booktitle = "Proceedings...",
                 year = "2004",
               editor = "Ara{\'u}jo, Arnaldo de Albuquerque and Comba, Jo{\~a}o Luiz Dihl 
                         and Navazo, Isabel and Sousa, Ant{\^o}nio Augusto de",
         organization = "Brazilian Symposium on Computer Graphics and Image Processing, 17. 
                         (SIBGRAPI) - Ibero-American Symposium on Computer Graphics, 2 
                         (SIACG)",
            publisher = "IEEE Computer Society",
              address = "Los Alamitos",
             keywords = "pattern recognition, multiple classifiers, handwritten 
                         recognition.",
             abstract = "This paper prsents a multiple classifier system applied to the 
                         handwritten word recognition (HWR) problem. The goal is to analyse 
                         the influence of different global classifiers taken isolatedly as 
                         well as combined in a particular HWR task. The application 
                         proposed is the recognition of the Portuguese handwritten names of 
                         the months. The strategy takes advantage of the complementarity 
                         mechanisms of three different classifiers: Conventional Neural 
                         Network, Class-Modular Neural Network and Hidden Markov Models, 
                         yielding a multiple classifier that is more efficient than either 
                         individual technique. The recognition rates obtained vary from 
                         75.9% using the stand alone HMM classifier to 96.0% considering 
                         the classifiers combination.",
  conference-location = "Curitiba, PR, Brazil",
      conference-year = "17-20 Oct. 2004",
                  doi = "10.1109/SIBGRA.2004.1352947",
                  url = "http://dx.doi.org/10.1109/SIBGRA.2004.1352947",
             language = "en",
                  ibi = "6qtX3pFwXQZeBBx/DcAw2",
                  url = "http://urlib.net/ibi/6qtX3pFwXQZeBBx/DcAw2",
           targetfile = "4382_oliveira_jose.pdf",
        urlaccessdate = "2024, May 01"
}