Olive fly sting detection based on computer vision

BEYAZ A., Martinez Gila D. M., Gomez Ortega J., Gamez Garcia J.

POSTHARVEST BIOLOGY AND TECHNOLOGY, cilt.150, ss.129-136, 2019 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 150
  • Basım Tarihi: 2019
  • Doi Numarası: 10.1016/j.postharvbio.2019.01.003
  • Dergi Adı: POSTHARVEST BIOLOGY AND TECHNOLOGY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.129-136
  • Anahtar Kelimeler: Olive fruit, Olea europaea, Olive fly, Bactrocera oleae, Olive fly sting, Image processing, Computer vision, VARIETIES, PEST
  • Ankara Üniversitesi Adresli: Evet

Özet

Olive fly (Bactrocera oleae Rossi) is a parasite that infects olive fruit by stinging the fruit to deposit its eggs. The infection damages the fruit during its growth and affects the quality of the olive oil as a final product. Detection of olive fly stings in the virgin olive oil production process is therefore critical. This research aimed to develop automatic methodologies based on computer vision to detect fly stings in infected olive fruit samples. Different methodologies to detect defective areas on the surface of the fruit and classify them between stings and other bruises are described. The methodology to detect defective areas reached a success ratio of 93% and best classification algorithms reached a success ratio close to 80%. The proposed methodology could be applied at reception of fruit for input of the virgin olive oil production process.