Linda Sutriani, Handaru Jati, Ketut Ima Ismara, Veny Betsy Saragih, Ali Impron
Palm oil is a critical plant in the vegetable oil industry, with Indonesia as one of its largest contributors. To enhance the management and monitoring of palm oil plantations efficiently, remote sensing technology using drones (Unmanned Aerial Vehicles, UAVs) is increasingly used. UAV aerial imagery allows direct monitoring of palm tree conditions; however, the main challenge is accurately detecting palm trees amid varying environmental conditions such as changing lighting and overlapping vegetation. Therefore, object detection algorithms like You Only Look Once (YOLO) are often applied in related studies. YOLO is known for its ability to detect objects directly with high precision, making it ideal for monitoring the growth and distribution of palm trees over vast plantation areas. Previous studies have shown that the application of YOLO for palm tree detection yields good performance, although the challenge of distinguishing between palm trees and other species like dates and coconuts remains a significant issue. This study aims to provide a comprehensive literature review on the application of the YOLO algorithm in palm tree identification and analyze current research trends and challenges. Through keyword network visualization analysis, some main topics include 'YOLO, RIPENESS, AND YOLO MODEL. The YOLO models used for detection are Yolov3, Yolov4, Yolov5, Yolov7, and Yolov8, which are not only for counting trees but also for classification. © 2025 IEEE.
Universitas Negeri Yogyakarta, Faculty of Information Tecnhnology, Yogyakarta, Indonesia; Universitas Muhammadiyah Sampit, Central Kalimantan, Indonesia; Universitas Muhammadiyah Sampit, Faculty of Engineering and Agriculture, Central Kalimantan, Indonesia
