Extended Random Jumps for Classification of Hyperspectral Data

Mouselli E., KASAPOĞLU N. G.

26th IEEE Signal Processing and Communications Applications Conference (SIU), İzmir, Türkiye, 2 - 05 Mayıs 2018, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu.2018.8404316
  • Basıldığı Şehir: İzmir
  • Basıldığı Ülke: Türkiye
  • İstanbul Kültür Üniversitesi Adresli: Hayır

Özet

Extended random walker (ERW) has shown a good performance in hyperspectral data classification. In this paper some adjustment on spatial relation is applied in ERW algorithm. Segments obtained from Watershed segmentation are used forming the new spatial space to improve the classification accuracy.