Endmember Detection using Enhanced Constrained Optimization in Hyperspectral Imaging

YÜKSEL ERDEM S. E.

22nd IEEE Signal Processing and Communications Applications Conference (SIU), Trabzon, Türkiye, 23 - 25 Nisan 2014, ss.1023-1026 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2014.6830406
  • Basıldığı Şehir: Trabzon
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.1023-1026
  • Hacettepe Üniversitesi Adresli: Evet

Özet

A new algorithm is presented for linear spectral mixture analysis that respects the constraints on the endmembers. The results show that it provides a more robust solution as compared to the ICE and SPICE algorithms due to the use of constrained quadratic optimization for endmember detection.