PlantVillage IoT & AI Platform

A full-stack smart agriculture platform combining IoT sensor monitoring, real-time dashboard, and AI-powered pest/disease detection.

---

Features

• IoT Sensors: ESP32-based sensors stream temperature, humidity, soil moisture, and water depth via MQTT.
• AI Detection: Upload or capture plant images for pest/disease detection using deep learning models.
• Real-Time Dashboard: Mobile app displays live sensor data and AI predictions.
• Socket.IO: Real-time communication between backend and frontend.
• REST API: Endpoints for disease and pest prediction.

---

Project Structure

elc_summer/
├── server/           # Flask backend, MQTT, AI models, database
├── my-expo-app/      # Expo React Native frontend app
├── mqtt/             # ESP32 Arduino sketch for sensors

---

Backend (server/)

• Tech: Flask, Flask-SocketIO, SQLAlchemy, Paho-MQTT, PyTorch, Timm
• Main entry: server/run.py
• Database: PostgreSQL

Setup

1. Install dependencies:

   cd server
   pip install -r requirements.txt

2. Configure environment variables in server/.env:

   DB_STRING = "postgresql://postgres:yourpassword@localhost:5432/Elc_summer"
   MQTT_USERNAME = "your_mqtt_user"
   MQTT_PASSWORD = "your_mqtt_pass"
   MQTT_HOST = "your_mqtt_host"
   MQTT_PORT = "8883"
   

3. Start PostgreSQL and create the database.
4. Run the server:

   python run.py

---

Frontend (my-expo-app/)

• Tech: React Native (Expo), NativeWind, Socket.IO client, Zustand
• Main entry: my-expo-app/App.tsx
• Pages: Home (dashboard), Manual (image upload/capture for AI detection)

Setup

1. Install dependencies:

   cd my-expo-app
   npm install

2. Set API base URL in my-expo-app/.env:

   API_BASE_URL = "http://<your-server-ip>:6000"
   

3. Start the app:

   npx expo start

---

ESP32 IoT (mqtt/)

• Flash the sketch in mqtt/sketch_jul09b/sketch_jul09b.ino to your ESP32.
• Update WiFi and MQTT credentials as needed.

---

API Endpoints

• POST /predict/disease — Upload image for disease detection.
• POST /predict/pest — Upload image for pest detection.

---

Useful Files

• server/run.py: Main backend entry point.
• server/app/ml/predictor.py: AI model inference.
• server/app/routes/routes.py: API routes.
• my-expo-app/src/pages/home.tsx: Dashboard UI.
• my-expo-app/src/pages/manual.tsx: Manual detection UI.

---

License

MIT License

---

For questions or contributions, open an