• DocumentCode
    2050095
  • Title

    Classification of DCT-based filtered multichannel images

  • Author

    Fevralev, Dmitriy ; Lukin, Vladimir ; Krivenko, Sergey ; Ponomarenko, Nikolay ; Kurekin, Andriy

  • Author_Institution
    Dept of Receivers, Transmitters & Signal Process., Nat. Aerosp. Univ., Kharkov, Ukraine
  • fYear
    2010
  • fDate
    23-27 Feb. 2010
  • Firstpage
    311
  • Lastpage
    311
  • Abstract
    Multichannel remote sensing (MRS) data can be passed to customers in different forms: original (raw), pre-filtered, compressed, classified. In this paper, we analyze how pre-filtering of original images can influence classification accuracy of three-channel images using three channels of real life Landsat TM data with simulated noise.
  • Keywords
    discrete cosine transforms; filtering theory; image classification; remote sensing; DCT-based filtering; filtered multichannel images; image classification; multichannel remote sensing data; three-channel images; Filtering; Hyperspectral imaging; Image analysis; Neural networks; PSNR; Remote monitoring; Remote sensing; Satellites; Support vector machine classification; Support vector machines; Filtering of multichannel images; classification;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET), 2010 International Conference on
  • Conference_Location
    Lviv-Slavske
  • Print_ISBN
    978-966-553-875-2
  • Electronic_ISBN
    978-966-553-901-8
  • Type

    conf

  • Filename
    5445891