Shape-Based Descriptor for Sunn Pest Damaged Wheat Kernel Detection

Kartal Y. , ÖZKAN K.

27th Signal Processing and Communications Applications Conference (SIU), Sivas, Türkiye, 24 - 26 Nisan 2019 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2019.8806333
  • Basıldığı Şehir: Sivas
  • Basıldığı Ülke: Türkiye


This study mainly focuses on the effect of sunn pest on wheat. The most prominent feature among the high quality wheat with sun-drenched wheat is the shape differences. For this reason, in this study, a shape recognition method based on the angle information of Fourier transformation is proposed. The performance of the proposed descriptor tested in data sets such as Mpeg-7, leaf, caltech 101, animal. In addition to these datasets, experimental studies were conducted in order to recognize non - classified wheat. Experimental results show that our proposed descriptor provides good accuracies indicating that Fourier Transform based local descriptor captures important characteristics of images that are useful for classification.