This repository contains implementations of the YOLOv5 architecture for specific agricultural use cases. The primary focus is the identification and localization of palm trees to support autonomous farming systems, such as automated fertilizing and pesticide spraying drones or robots.
- Palm Tree Detection: Trained models specifically optimized to identify palm trees from aerial or ground-level imagery.
- Multi-Export Support: Inference-ready code for PyTorch, ONNX, TFLite, and CoreML formats.
- Automation Integration: Designed to provide coordinate data for automated spraying and fertilizing systems.
- High Performance: Real-time detection capabilities suitable for deployment on edge devices (Jetson Nano, Raspberry Pi, etc.).
- Architecture: YOLOv5 (You Only Look Once v5)
- Framework: PyTorch
- Inference Engines: ONNX, TensorFlow Lite, CoreML
- Focus: Agrotech, Computer Vision, Precision Farming
- Target Objects: Palm Trees (Kelapa Sawit)
Clone the Project .. code-block:: bash
git clone https://github.com/afafirmansyah/object-detection-yolov5.git
Environment Setup - It is recommended to use a virtual environment or Conda. - Install the required dependencies:
pip install -r requirements.txt
Running Detection - To run inference using the pre-trained model on your images or videos:
python detect.py --weights best.pt --source path/to/your/images/
Automation Output - The system generates bounding box coordinates that can be sent to a controller (via MQTT or Serial) for automated spraying actions.
[Tempatkan gambar hasil deteksi pohon sawit dengan bounding boxes di sini]
This project is licensed under the MIT License - see the license.txt file for details.
Ahmad Fauzi Firmansyah - GitHub: afafirmansyah - LinkedIn: ahmad-fauzi-firmansyah