CoffeeSentinel-X is an AI-powered mobile system built on ODK-X for diagnosing Coffee Wilt Disease (CWD) in real time and supporting smallholder coffee farmers with decision tools and agronomic advice.
- 🔍 Project Objectives
- 🌿 Key Features
- 🧠 Technical Stack
- 📂 Repository Structure
- 🚀 Getting Started
- 🏗️ Model Overview
- 🌍 Use Cases
- 🤝 Contributing
- 📄 License
- 📅 Project Status
- 🌟 Acknowledgements
- 📬 Contact
- ✅ Curate and annotate a dataset of coffee plants (healthy and diseased)
- ✅ Train a CNN-based image model for CWD detection
- ✅ Integrate model with ODK-X for mobile usage
- ✅ Extend with features useful to coffee farmers (mapping, advice, tracking)
- 📷 Camera-based Diagnosis: Detects Coffee Wilt Disease from plant images.
- 🧠 On-device ML Model: Efficient TensorFlow Lite implementation.
- 📍 GPS Logging: Location-aware plot and plant tracking.
- 🔁 Offline-first Workflow: Designed for rural areas with low connectivity.
- 📋 ODK-X Forms: Survey-based task and data collection.
- 🌱 Farmer Support Tools: Guidance, logging, and market tips (in development).
| Component | Technology |
|---|---|
| Survey Interface | ODK-X (Survey, Services) |
| ML Inference | TensorFlow Lite |
| Annotation | CVAT or Roboflow |
| Backend (optional) | Django + PostgreSQL |
| Deployment | Android APK, Firebase |
| Hosting | GitHub |
📂 Repository Structure
CoffeeSentinel-X/ │ ├── model/ # Training code, notebooks, TFLite model │ ├── train/ │ └── tflite/ │ ├── odkx_app/ # ODK-X forms and sync config │ ├── survey/ │ └── services/ │ ├── dashboard/ # Optional backend/visualization layer │ ├── backend/ │ └── frontend/ │ ├── docs/ # Documentation files │ ├── MODEL_ARCHITECTURE.md │ ├── DATA_CURATION.md │ └── USER_GUIDE.md │ ├── media/ # Sample images, logos, visuals │ ├── LICENSE └── README.mdyaml Copy Edit