@InProceedings{Julca-AguilarMaiaHira:2017:TeClCo,
               author = "Julca-Aguilar, Frank Dennis and Maia, Ana Lucia Lima Marreiros and 
                         Hirata, Nina Sumiko Tomita",
          affiliation = "{University of S{\~a}o Paulo} and State University of Feira de 
                         Santana, University of S{\~a}o Paulo and {University of S{\~a}o 
                         Paulo}",
                title = "Text/non-text classification of connected components in document 
                         images",
            booktitle = "Proceedings...",
                 year = "2017",
               editor = "Torchelsen, Rafael Piccin and Nascimento, Erickson Rangel do and 
                         Panozzo, Daniele and Liu, Zicheng and Farias, Myl{\`e}ne and 
                         Viera, Thales and Sacht, Leonardo and Ferreira, Nivan and Comba, 
                         Jo{\~a}o Luiz Dihl and Hirata, Nina and Schiavon Porto, Marcelo 
                         and Vital, Creto and Pagot, Christian Azambuja and Petronetto, 
                         Fabiano and Clua, Esteban and Cardeal, Fl{\'a}vio",
         organization = "Conference on Graphics, Patterns and Images, 30. (SIBGRAPI)",
            publisher = "IEEE Computer Society",
              address = "Los Alamitos",
             keywords = "text segmentation, connected component, convolutional neural 
                         network.",
             abstract = "Text segmentation is an important problem in document analysis 
                         related applications. We address the problem of classifying 
                         connected components of a document image as text or non-text. 
                         Inspired from previous works in the literature, besides common 
                         size and shape related features extracted from the components, we 
                         also consider component images, without and with context 
                         information, as inputs of the classifiers. Muli-layer perceptrons 
                         and convolutional neural networks are used to classify the 
                         components. High precision and recall is obtained with respect to 
                         both text and non-text components.",
  conference-location = "Niter{\'o}i, RJ, Brazil",
      conference-year = "17-20 Oct. 2017",
                  doi = "10.1109/SIBGRAPI.2017.66",
                  url = "http://dx.doi.org/10.1109/SIBGRAPI.2017.66",
             language = "en",
                  ibi = "8JMKD3MGPAW/3PFS8CH",
                  url = "http://urlib.net/ibi/8JMKD3MGPAW/3PFS8CH",
           targetfile = "PID4960469.pdf",
        urlaccessdate = "2024, Apr. 27"
}