Halyomorpha Halys, commonly known as the brown marmorated stink bug, is an invasive insect that causes significant damage in orchards. Neural networks have the potential to improve insect pest detection and classification in modern agriculture, which can lead to better pest management. The detection of these insects in orchards using drones imposes special problems because the images are taken from a limited distance and the foliage of the trees makes detection difficult. In this article, we studied the possibility of detecting the respective insects using the latest generation YOLOv8 neural networks and compared the results with the well-known YOLOv5 network. The results were obviously better for YOLOv8 (accuracy of 94.55%). However, satisfactory results were also obtained in the case of YOLOv5 (accuracy of 90.91%).

Halyomorpha Halys Detection in Orchard from UAV Images Using Convolutional Neural Networks / Dinca A.; Popescu D.; Pinotti C.M.; Ichim L.; Palazzetti L.; Angelescu N.. - ELETTRONICO. - 14135 LNCS:(2023), pp. 315-326. (Intervento presentato al convegno International Work-Conference on Artificial Neural Networks IWANN 2023: Advances in Computational Intelligence) [10.1007/978-3-031-43078-7_26].

Halyomorpha Halys Detection in Orchard from UAV Images Using Convolutional Neural Networks

Palazzetti L.;
2023

Abstract

Halyomorpha Halys, commonly known as the brown marmorated stink bug, is an invasive insect that causes significant damage in orchards. Neural networks have the potential to improve insect pest detection and classification in modern agriculture, which can lead to better pest management. The detection of these insects in orchards using drones imposes special problems because the images are taken from a limited distance and the foliage of the trees makes detection difficult. In this article, we studied the possibility of detecting the respective insects using the latest generation YOLOv8 neural networks and compared the results with the well-known YOLOv5 network. The results were obviously better for YOLOv8 (accuracy of 94.55%). However, satisfactory results were also obtained in the case of YOLOv5 (accuracy of 90.91%).
2023
International Work-Conference on Artificial Neural Networks
International Work-Conference on Artificial Neural Networks IWANN 2023: Advances in Computational Intelligence
Dinca A.; Popescu D.; Pinotti C.M.; Ichim L.; Palazzetti L.; Angelescu N.
File in questo prodotto:
File Dimensione Formato  
HH-Detection-in-Orchard-from-UAV-Images-Using-Convolutional-Neural-Networks.pdf

Accesso chiuso

Tipologia: Preprint (Submitted version)
Licenza: DRM non definito
Dimensione 820.55 kB
Formato Adobe PDF
820.55 kB Adobe PDF   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1347492
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact