Description of Invention
This research investigates the use of deep learning models to detect mango fruits from aerial images captured by drones, focusing on Faster R-CNN and YOLO variants. The study evaluates these models using datasets from the University of Sydney, CQUniversity, and a locally prepared dataset. Among the models, YOLOv8 demonstrated the highest performance, achieving 98.5% mAP on the Sydney dataset and 80.9% mAP on the local dataset. Known for its real-time capability, YOLOv8 also shows strong potential for applications requiring rapid detection. The MangoVision GUI, designed with bilingual support and features such as GPS coordinate extraction, integrates these models into a user-friendly tool for precise mango detection. This study highlights the potential of AI-driven solutions to enhance agricultural practices, making YOLOv8 a top choice for precision farming.