Machine Learning and IOT-Based Paddy Rice Monitoring System

Award: TOP 1 Best Paper – Computing and Information Systems Category
Research In-House 2025 | Bohol Island State University - Main Campus

A machine learning and IoT-based paddy rice post-drying monitoring system that integrates real-time sensor data with predictive analytics to optimize rice yield tracking and moisture management across municipal, barangay, and farmer levels.

---

Overview

Paddy Rice Tracker is an end-to-end IoT-enabled agricultural monitoring platform designed to: -Capture real-time environmental data (temperature, humidity, moisture) from ESP32/Arduino sensors -Predict post-drying yield using machine learning models -Provide role-based dashboards for municipal officers, barangay staff, and farmers -Provide automated shelf-life recommendations based on final moisture content to optimize storage and reduce spoilage -Synchronize data across edge devices via RESTful APIs and Cloud Database storage -Generate temporal analytics to drive data-informed agricultural decisions

---

Hardware Component

The system relies on edge devices to collect real-time sensor data. The source code for the hardware logic is maintained in a separate repository to decouple the web logic from the embedded systems.

• Repository: paddy-rice-iot-node
• Role: Handles sensor data collection (Arduino), runs local ML predictions, and syncs data to this web platform via the /api/sync endpoint.

---

Key Features

Role-Based Access Control (RBAC)

The system supports three distinct user tiers with specific privileges:

1. Municipal Officer:
   • View aggregated yield statistics across all Barangays.
   • Monitor high-level agricultural performance.
2. Barangay Staff:
   • Manage local farmer registries.
   • Validate and oversee drying records for their jurisdiction.
   • View barangay-specific analytics.
3. Farmer:
   • Personal dashboard to track their specific batch records.
   • View historical drying performance and output.

Real-Time Analytics Dashboards

• Interactive Chart.js visualizations for yield trends (monthly/yearly)
• Chronologically sorted time-series data with proper year-month ordering
• Comparative analysis of initial vs. final weight per batch and barangay
• Tracking of temperature, humidity, and moisture levels for every drying batch

IoT Data Pipeline

• RESTful endpoints for secure sensor data ingestion from ESP32/Arduino devices
• UUID-based record tagging for audit trails and duplicate prevention
• JSON-based data syncing between edge devices and the cloud database
• Server-side calculation of moisture reduction and final yield upon data receipt

Data Management

• CRUD operations for drying records, farmers, and locations
• PostgreSQL database architecture managed via SQLAlchemy ORM
• Enforced foreign key constraints across Municipalities → Barangays → Farmers → Records

The Team

• Jennifer Tongco – Team Leader, Lead Full-Stack Developer & System Architect
• Claire Justin Pugio – UI/UX Designer
• John Jabez Visarra – Hardware Engineer & Frontend Support
• John Kylo Cubelo – Hardware Engineer
• Engr. Jeralyn Alagon – Thesis Adviser

API Endpoints

Endpoint	Method	Description
/api/sync	POST	Sync drying records from IoT devices
/api/fetch?farmer_uuid=<uuid>	GET	Retrieve farmer's historical records
/api/farmers/<username>	GET	Fetch farmer profile by username
/api/users	GET	List all users (municipal/barangay)
/api/barangays	GET	List all barangays
/api/municipalities	GET	List all municipalities

---

Tech Stack

Backend

• Framework: Flask 2.3.2 (Python)
• Database: PostgreSQL (production) / SQLite (local dev)
• ORM: SQLAlchemy 3.1.1
• Migrations: Alembic (Flask-Migrate)
• Authentication: Flask-Login, Flask-Dance (OAuth), Werkzeug

Frontend

• Templating: Jinja2
• Visualization: Chart.js (Yield trends & analytics)
• Styling: Bootstrap 5 (Responsive UI)

IoT & ML Stack (Inferred from Data Schema)

• Microcontroller: Raspberry Pi 4 / Arduino (Sensor integration)
• ML Libraries: Scikit-learn 1.3.2, Joblib, NumPy, Pandas
• Algorithm: Regression Models (trained for moisture/yield prediction)
• Data Synchronization: RESTful API (Custom JSON sync via requests)

Deployment

• Platform: Render.com
• WSGI Server: Gunicorn 21.2.0
• Environment: Python 3.11

---

Project Structure

paddy_rice_tracker/
├── app.py                      # Flask application entry point
├── requirements.txt            # Python dependencies
├── Procfile                    # Render deployment config
├── render.yaml                 # Render service definition
├── website/
│   ├── __init__.py             # Flask app factory
│   ├── models.py               # SQLAlchemy models (User, Farmer, DryingRecord, etc.)
│   ├── views.py                # Main routes (dashboards, CRUD)
│   ├── api.py                  # RESTful API endpoints
│   ├── auth.py                 # Authentication & user management
│   ├── extensions.py           # Flask extensions (db, login_manager, migrate)
│   ├── static/
│   │   ├── logo.svg
│   │   ├── favicon.svg
│   │   └── index.js
│   └── templates/
│       ├── base.html           # Base template with navbar
│       ├── login.html
│       ├── dashboard.html      # Role-specific dashboards
│       ├── analytics.html      # Municipal analytics
│       ├── barangay_analytics.html
│       ├── farmer_analytics.html
│       ├── records.html
│       ├── farmers.html
│       └── ...
├── migrations/                 # Alembic database migrations
└── instance/
    └── database.db             # Local SQLite database (dev only)

---

Getting Started

Prerequisites

• Python 3.11+
• PostgreSQL (for production) or SQLite (auto-created for dev)
• pip package manager

Installation

1. Clone the repository

   git clone https://github.com/jenn-if-err/paddy_rice_tracker.git
   cd paddy_rice_tracker

2. Create a virtual environment

   python -m venv venv
   # Windows
   venv\Scripts\activate
   # macOS/Linux
   source venv/bin/activate

3. Install dependencies

   pip install -r requirements.txt

4. Set up environment variables

   Create a .env file in the root directory:

   SECRET_KEY=your_secret_key_here
   DATABASE_URL=sqlite:///instance/database.db  # For local dev
   # DATABASE_URL=postgresql://user:password@host/dbname  # For production
   GOOGLE_CLIENT_ID=your_google_oauth_client_id
   GOOGLE_CLIENT_SECRET=your_google_oauth_client_secret

5. Initialize the database

   flask db init
   flask db migrate -m "Initial migration"
   flask db upgrade

6. Run the development server

   python app.py

   Navigate to http://127.0.0.1:5000

---

Database Schema

Core Models

User (Municipal & Barangay Staff)

• id, email, full_name, role (municipal/barangay)
• municipality_id, barangay_id, password

Farmer

• id, uuid, first_name, middle_name, last_name
• username, password, barangay_id, user_id

DryingRecord

• id, uuid, timestamp, batch_name
• initial_weight, final_weight, temperature, humidity
• sensor_value, initial_moisture, final_moisture, drying_time
• date_dried, date_planted, date_harvested, due_date
• farmer_id, barangay_id, municipality_id, user_id

Municipality & Barangay

• Hierarchical location management

---

API Usage Examples

Sync IoT Data

POST /api/sync
Content-Type: application/json

{
  "records": [
    {
      "uuid": "123e4567-e89b-12d3-a456-426614174000",
      "batch_name": "Batch A",
      "initial_weight": 100.5,
      "temperature": 28.5,
      "humidity": 65.2,
      "sensor_value": 320,
      "initial_moisture": 24.5,
      "final_moisture": 14.0,
      "drying_time": "48 hours",
      "final_weight": 85.3,
      "date_dried": "2025-12-01",
      "farmer_uuid": "farmer-uuid-here",
      "user_id": 1
    }
  ]
}

Fetch Farmer Records

GET /api/fetch?farmer_uuid=farmer-uuid-here

Response:

[
  {
    "uuid": "...",
    "batch_name": "Batch A",
    "initial_weight": 100.5,
    "final_weight": 85.3,
    "temperature": 28.5,
    "humidity": 65.2,
    "date_dried": "2025-12-01",
    "farmer_name": "Juan Dela Cruz",
    "barangay_name": "San Jose",
    "municipality_name": "Cabanatuan City"
  }
]

---

Deployment

Render.com (Current Deployment)

1. Connect GitHub repository to Render dashboard
2. Environment variables are auto-configured via render.yaml
3. Database is provisioned as a managed PostgreSQL instance
4. Auto-deploy on push to main branch

Manual Deployment (Alternative)

# Install production dependencies
pip install -r requirements.txt

# Set production environment variables
export FLASK_ENV=production
export DATABASE_URL=postgresql://...

# Run database migrations
flask db upgrade

# Start Gunicorn server
gunicorn app:app --bind 0.0.0.0:8000

---

Testing

Run Local Development Server

python app.py

Debug mode is enabled by default in app.py for hot-reloading.

Test API Endpoints

# Test sync endpoint
curl -X POST http://127.0.0.1:5000/api/sync \
  -H "Content-Type: application/json" \
  -d '{"records": [...]}'

# Test fetch endpoint
curl http://127.0.0.1:5000/api/fetch?farmer_uuid=<uuid>

---

License

This project is part of an academic thesis and is intended for educational purposes.

---