Plant Counting By Using k-NN Classification on UAVs Images

Tavus M. R., Eker M. E., Senyer N., Karabulut B.

23nd Signal Processing and Communications Applications Conference (SIU), Malatya, Turkey, 16 - 19 May 2015, pp.1058-1061 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu.2015.7130015
  • City: Malatya
  • Country: Turkey
  • Page Numbers: pp.1058-1061
  • Keywords: k-NN algorithm, plant counting, image processing
  • Ondokuz Mayıs University Affiliated: Yes


In this study, Plant Counting was implemented by appling k-NN classification to images obtained from Unnamed Air Vehicle (UAV). Firstly, The images were subjected to erosion process by transforming different colour levels. The objects in the images were classified as plant and soil by means of k-NN classification. It was observed that plants can be counted with 87,7% of accuracy and 86,6% of precision by being performed last processing of the morphology of the binary image.