This application utilizes YOLOv11, a state-of-the-art object detection algorithm, to detect pests and diseases on tomato leaves. The model was trained on a large dataset of tomato leaf images and achieves impressive results.
The YOLOv11 model was evaluated on a test dataset after 50 epochs with a batch size of 100. The evaluation metrics are:
- Precision: 0.3671
- Recall: 0.2634
- mAP@50: 0.2814
- mAP@50-95: 0.1429
- Fitness: 0.1567
These metrics demonstrate the model's ability to accurately detect pests and diseases on tomato leaves.
- Real-time detection: Detect pests and diseases in real-time using the YOLOv11 algorithm.
- High accuracy: Achieves high precision and recall rates.
- Tomato leaf dataset: Trained on a large dataset of tomato leaf images.
- User-friendly interface: Easy-to-use interface for uploading images and viewing detection results.
- Python 3.12
- Streamlit
- Ultralytics YOLOv11
- OpenCV
- Clone the repository.
- Install required dependencies in the requirements.txt file
- Run the application using streamlit using streamlit run app.py commad on your linux ternimal.
- Upload an image of a tomato leaf.
- View detection results.
Future Development
- Improve model performance by increasing dataset size and training epochs.
- Integrate with other computer vision algorithms for enhanced detection.
- Develop a mobile application for farmers and agricultural professionals.