Resumen
Tree counting is an important plantation practice for biological asset inventories, etc. The application of precision agriculture in counting oil palm trees can be implemented by detecting oil palm trees from aerial imagery. This research uses the deep learning approach using YOLOv3, YOLOv4, and YOLOv5m in detecting oil palm trees. The dataset consists of drone images of an oil palm plantation acquired using a Fixed Wing VTOL drone with a resolution of 5cm/pixel, covering an area of 730 ha labeled with an oil palm class of 56,614 labels. The test dataset covers an area of 180 ha with flat and hilly conditions with sparse, dense, and overlapping canopy and oil palm trees intersecting with other vegetations. Model testing using images from 24 regions, each of which covering 12 ha with up to 1000 trees (for a total of 17,343 oil palm trees), yielded F1-scores of 97.28%, 97.74%, and 94.94%, with an average detection time of 43 s, 45 s, and 21 s for models trained with YOLOv3, YOLOv4, and YOLOv5m, respectively. This result shows that the method is sufficiently accurate and efficient in detecting oil palm trees and has the potential to be implemented in commercial applications for plantation companies.