Identification of sunn pest (Eurygaster) in cereal farms using UAV image processing technique

Shadmanfar, Somaye; Lotfalian Dehkordi, Amin; Ghasemi Dastgerdi, Ahmad Reza; Haghshenas Nasrabadi, Alireza

doi:10.22034/jrmam.2023.13941.596

Identification of sunn pest (Eurygaster) in cereal farms using UAV image processing technique

Document Type : Original Article

Authors

somaye Shadmanfar ¹

amin lotfalian dehkordi ²

ahmad reza Ghasemi Dastgerdi ³

Alireza Haghshenas Nasrabadi ⁴

¹ M.Sc., Department of Mechanical Engineering of Biosystems, Faculty of Agriculture, Shahrekord University, Shahrekord, Iran.

² Assistant Professor, Department of Mechanical Engineering of Biosystems, Faculty of Agriculture, Shahrekord University, Shahrekord, Iran.

³ Associate Professor, Water Engineering Department, Faculty of Agriculture, Shahrekord University, Shahrekord, Iran.

⁴ Lecturer, Plant Protection Research Department, Isfahan Agricultural & Natural Resources Research Center, Isfahan, Iran

10.22034/jrmam.2023.13941.596

Abstract

Considering that wheat and barley are important and strategic products in Iran, the Sunn pest is one of the most important pests of these products. Every year, the country's plant protection organization monitors this dangerous pest so that an order to fight it can be issued at the right time. Farm monitoring is a very difficult, time-consuming, and destructive task; it also requires a large number of specialized human resources. One of the goals of this research is to create a non-destructive method for such an operation in the identification of Sunn pests. In this research, the counting of Sunn pests in cereal farms was done by two methods using aerial image processing by recording 954 images and framing method and eye counting in the farm by an expert in two farms with different numbers of this pest. The number of pests in each image was obtained after processing the images using the Python programming language in the PyCharm learning library. The statistical study results showed no significant difference at the 1% probability level between the average data of the two methods. This method can be used confidently to monitor the Sunn pest in cereal farms. In this research, by using UAV image processing, a fast, accurate, and non-destructive method was proposed and evaluated for counting the pests and issuing spraying orders in cereal farms.

Keywords

Image processing

UAV

Sunn pest

Eurygaster counting

Machine learning

Subjects