This paper presents the results of a case study focusing on automating the monitoring process of Halyomorpha halys (HH) in smart agriculture. HH is an invasive global pest that causes significant economic damages to fruit orchards. Our study aims to address the challenges associated with HH scouting, which is traditionally a time-and labor-intensive task. The study concentrates on HALY. ID project achievements of 2022 campaign, where the image acquisition is performed using solely a drone, and exploiting an autonomous drone-based navigation protocol from the top of the orchard. The protocol for the drone involves flying through predefined waypoints and capturing pictures at various tree positions. Moreover, for the sake of simplicity we conducted the detection of HH class only. We performed analyses on the acquired images, including evaluations of both image blurriness and brightness. Then, we obtain encouraging results from YOLOV5 detection algorithms trained on the novel acquired dataset of images. These outcomes show promising potential for automating HH monitoring and mark a significant step towards enhancing smart agriculture practices.
A Drone-based Automated Halyomorpha halys Scouting: A Case Study on Orchard Monitoring / Francesco Betti Sorbelli, Lorenzo Palazzetti, Cristina M Pinotti. - ELETTRONICO. - (2023), pp. 0-0. (Intervento presentato al convegno 2023 IEEE INTERNATIONAL WORKSHOP ON Metrology for Agriculture and Forestry) [10.1109/MetroAgriFor58484.2023.10424287].
A Drone-based Automated Halyomorpha halys Scouting: A Case Study on Orchard Monitoring
Francesco Betti Sorbelli;Lorenzo Palazzetti
;
2023
Abstract
This paper presents the results of a case study focusing on automating the monitoring process of Halyomorpha halys (HH) in smart agriculture. HH is an invasive global pest that causes significant economic damages to fruit orchards. Our study aims to address the challenges associated with HH scouting, which is traditionally a time-and labor-intensive task. The study concentrates on HALY. ID project achievements of 2022 campaign, where the image acquisition is performed using solely a drone, and exploiting an autonomous drone-based navigation protocol from the top of the orchard. The protocol for the drone involves flying through predefined waypoints and capturing pictures at various tree positions. Moreover, for the sake of simplicity we conducted the detection of HH class only. We performed analyses on the acquired images, including evaluations of both image blurriness and brightness. Then, we obtain encouraging results from YOLOV5 detection algorithms trained on the novel acquired dataset of images. These outcomes show promising potential for automating HH monitoring and mark a significant step towards enhancing smart agriculture practices.File | Dimensione | Formato | |
---|---|---|---|
MetroAgriFor_2023___HH_detection_via_ML_22.pdf
Accesso chiuso
Tipologia:
Preprint (Submitted version)
Licenza:
Tutti i diritti riservati
Dimensione
3.17 MB
Formato
Adobe PDF
|
3.17 MB | Adobe PDF | Richiedi una copia |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.