A Probabilistic Method for the Classification of Hyperspectral Images

Kutluk S., KAYABOL K., Akan A.

24th Signal Processing and Communication Application Conference (SIU), Zonguldak, Turkey, 16 - 19 May 2016, pp.905-908, (Full Text) identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume:
  • Doi Number: 10.1109/siu.2016.7495887
  • City: Zonguldak
  • Country: Turkey
  • Page Numbers: pp.905-908
  • Istanbul University-Cerrahpasa Affiliated: No

Abstract

In this study a supervised classification and dimensionality reduction method for hyperspectral images is proposed. For this purpose, using probabilistic principal component analysis (PPCA), dimensionality reduction is performed and a Gaussian mixture model (GMM) is built. Alongside this mixture model, spatial information is also included into the classification process by taking advantage of pixel neighborhoods.