@InProceedings{PapaFalMirSuzMas:2007:DeRoPa,
               author = "Papa, Jo{\~a}o Paulo and Falc{\~a}o, Alexandre X. and Miranda, 
                         Paulo A. V. and Suzuki, Celso T. N. and Mascarenhas, Nelson D. 
                         A.",
          affiliation = "LIV-IC-UNICAMP and LIV-IC-UNICAMP and LIV-IC-UNICAMP and 
                         LIV-IC-UNICAMP and Dept. of Computing, UFSCar",
                title = "Design of robust pattern classifiers based on optimum-path 
                         forests",
            booktitle = "Proceedings...",
                 year = "2007",
               editor = "Banon, Gerald Jean Francis and Barrera, Junior and Braga-Neto, 
                         Ulisses de Mendon{\c{c}}a and Hirata, Nina Sumiko Tomita",
                pages = "337--348",
         organization = "International Symposium on Mathematical Morphology, 8. (ISMM).",
            publisher = "Instituto Nacional de Pesquisas Espaciais (INPE)",
              address = "S{\~a}o Jos{\'e} dos Campos",
             keywords = "supervised classifiers, image foresting transform, image analysis, 
                         pattern recognition.",
             abstract = "We present a supervised pattern classifier based on optimum path 
                         forest. The samples in a training set are nodes of a complete 
                         graph, whose arcs are weighted by the distances between sample 
                         feature vectors. The training builds a classifier from key samples 
                         (prototypes) of all classes, where each prototype defines an 
                         optimum path tree whose nodes are its strongest connected samples. 
                         The optimum paths are also considered to label unseen test samples 
                         with the classes of their strongest connected prototypes. We show 
                         how to find prototypes with none classification errors in the 
                         training set and propose a learning algorithm to improve accuracy 
                         over an evaluation set. The method is robust to outliers, handles 
                         non-separable classes, and can outperform support vector 
                         machines.",
  conference-location = "Rio de Janeiro",
      conference-year = "October 10-13, 2007",
                 isbn = "978-85-17-00035-5",
             language = "en",
         organisation = "Universidade de S{\~a}o Paulo (USP)",
                  ibi = "83LX3pFwXQZ3qyBY/PKn22",
                  url = "http://urlib.net/ibi/83LX3pFwXQZ3qyBY/PKn22",
           targetfile = "ISMM2007fullpaper/fullpaper.tex",
                 type = "Watershed segmentation",
               volume = "1",
        urlaccessdate = "16 jun. 2024"
}