@InProceedings{TavaresSant:2017:ExInSt,
               author = "Tavares, Eduardo de Ara{\'u}jo and dos Santos, Jefersson Alex",
          affiliation = "{Universidade Federal de Minas Gerais} and {Universidade Federal 
                         de Minas Gerais}",
                title = "Exploiting indexing structures for large scale Remote Sensing 
                         Image Classification",
            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 = "remote sensing, image indexing.",
             abstract = "The rapid increase on the volume of data generated by remote 
                         sensing systems boosted by the evolution of satellites and the 
                         popularization of their imagery has enabled a wide range of new 
                         Earth Observation applications. At the same time, it created the 
                         challenge of how to efficiently deal with these collections of 
                         data. In this work we evaluate the use of indexing techniques for 
                         speeding up remote sensing image retrieval aiming automatic large 
                         scale geographical mapping in the future. Three CNNs are employed 
                         as feature extractors and compared to three low-level features on 
                         retrieval tasks performed on a dataset of aerial images with the 
                         LSH algorithm. Preliminary results showed a recall level of almost 
                         50% when only roughly 5% of the samples of the evaluated dataset 
                         needed to be considered.",
  conference-location = "Niter{\'o}i, RJ, Brazil",
      conference-year = "17-20 Oct. 2017",
             language = "en",
                  ibi = "8JMKD3MGPAW/3PK8MLE",
                  url = "http://urlib.net/ibi/8JMKD3MGPAW/3PK8MLE",
           targetfile = "2017_sibgrapi camera ready.pdf",
        urlaccessdate = "2024, Apr. 27"
}