In precision agriculture, adopting innovative methods is essential to optimize resource management practices, such as irrigation, fertilization, and disease control, particularly in areas requiring targeted interventions. Accurate identification and mapping of individual trees are critical for efficient farm management, enabling us to monitor tree growth in real-time and implement precision agriculture techniques effectively. In this study, we introduce a novel method for the automated detection and geolocation of olive tree crowns in large-scale plantations using deep learning and aerial imagery. We employed drones, specifically UAVs (Unmanned Aerial Vehicles), to capture high-resolution images, providing comprehensive coverage of agricultural zones in Meknes, Morocco. These images were processed using four YOLOv8 variants (n, s, l, and m), each selected for its balance between speed, computational efficiency, and accuracy. YOLOv8-l achieved the highest precision of 0.990 and recall of 0.973 with an F1 score of 0.981 and a mAP@50 of 0.994. By integrating image metadata with photogrammetry principles, we refined GPS (Global Positioning System) coordinates for each tree, allowing precise mapping of their locations. This method improves crop management, supports modern agriculture, and enables accurate mapping for better decisions.