@InProceedings{SilvaMontHiraHira:2017:ImOpLe,
               author = "Silva, Augusto C{\'e}sar Monteiro and Montagner, Igor dos Santos 
                         and Hirata Jr, Roberto and Hirata, Nina Sumiko Tomita",
          affiliation = "{Institute of Mathematics and Statistics} and {Institute of 
                         Mathematics and Statistics} and {Institute of Mathematics and 
                         Statistics} and {Institute of Mathematics and Statistics}",
                title = "Image operator learning based on local features",
            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 = "Sociedade Brasileira de Computa{\c{c}}{\~a}o",
              address = "Porto Alegre",
             keywords = "morphological operators, local features, image operator 
                         learning.",
             abstract = "Morphological operators in image processing have a wide range of 
                         applications, like in medical imaging and document image analysis. 
                         The design of such operators are made, mainly, by a trial and 
                         error approach. Another method to design these operators consists 
                         in using machine learning algorithms to define a local 
                         transformation that represents an operator. Previous works used 
                         mainly the intensity values of the pixels as feature vectors in 
                         the machine learning algorithms. We propose to extract different 
                         features, calculated from the image, to create different feature 
                         vectors to be used in the machine learning algorithms. We 
                         experiment this approach in four different public datasets, and 
                         results show that different features have a significant impact on 
                         the learned operators, but, just like the operators, the feature 
                         that provides better results also depends on the dataset used.",
  conference-location = "Niter{\'o}i, RJ, Brazil",
      conference-year = "17-20 Oct. 2017",
             language = "en",
                  ibi = "8JMKD3MGPAW/3PJ5ECP",
                  url = "http://urlib.net/ibi/8JMKD3MGPAW/3PJ5ECP",
           targetfile = "image-operator-learning-camera-ready.pdf",
        urlaccessdate = "2024, May 02"
}