Agriculture is the backbone of many economies, yet plant diseases continue to cause massive crop losses every year. What if farmers could detect diseases instantly using just an image?

That’s exactly what this project does 👇

🚀 Overview

The Plant Disease Detection System is an AI-powered web application that leverages Deep Learning (CNN) to identify plant diseases from leaf images. Along with detection, it also provides fertilizer recommendations, treatment steps, and prevention guidance.

Built using TensorFlow + Django, this project bridges the gap between AI and real-world agriculture 🌾

🎯 Key Features

• 🔍 Real-time Disease Detection
  Upload a plant leaf image and get instant predictions

• 🎯 High Accuracy
  Achieves 81–84% accuracy across 38 disease classes

• 💊 Smart Recommendations
  Get treatment steps, fertilizers, and prevention tips

• 📊 Prediction History
  Track all predictions with timestamps

• 📱 Responsive UI
  Works smoothly on desktop, tablet, and mobile

• 🚀 REST API Support
  Easy integration with mobile apps

• 🔐 Secure System
  CSRF protection, file validation, and sanitized inputs

🌾 Supported Crops & Diseases

The model supports 38 disease classes across multiple crops:

• 🍅 Tomato (10 classes)
• 🥔 Potato (3 classes)
• 🌽 Corn (4 classes)
• 🌶️ Pepper (2 classes)
• 🍎 Additional crops included

📸 Application Workflow

1. Upload a leaf image
2. Model analyzes using CNN
3. Returns:

• Disease name
• Confidence score
• Treatment
• Fertilizer advice
• Prevention tips

🚀 Quick Start

🔧 Prerequisites

• Python 3.8+
• pip
• Virtual environment (recommended)
• 4GB RAM
• 2GB storage

⚙️ Installation

# Clone repo
git clone https://github.com/dassomnath99/Plant-Disease-Detection.git
cd Plant-Disease-Detection

# Create virtual environment
python -m venv venv

# Activate (Windows)
venv\Scripts\activate

# Activate (Linux/macOS)
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

🤖 Model Setup

Option A: Use Pre-trained Model

Place these files in /models:

• plant_disease_model.h5
• class_names.json

Option B: Train Your Own Model

python download_data.py
python train_with_test.py

🌐 Run Django Server

python manage.py makemigrations
python manage.py migrate
python manage.py runserver

Open 👉 http://127.0.0.1:8000/

🔌 API Example

Predict Disease

POST /api/predict/

Response:

{
  "success": true,
  "prediction": {
    "disease": "Tomato_Late_blight",
    "confidence": 94.32,
    "plant_type": "Tomato",
    "treatment": "Apply fungicide",
    "prevention": "Avoid wet foliage"
  }
}

🧪 Model Performance

📊 Training Details

• Dataset: PlantVillage (54K+ images)
• Architecture: MobileNetV2 (Transfer Learning)
• Framework: TensorFlow 2.15
• Input Size: 224x224
• Model Size: ~13MB

🛠️ Tech Stack

Backend

• Django
• Django REST Framework
• TensorFlow / Keras

Frontend

• HTML5
• CSS3
• JavaScript

ML Techniques

• CNN (MobileNetV2)
• Data Augmentation
• Adam Optimizer

📱 Mobile Integration

Works seamlessly with:

• React Native
• Flutter

You can easily send images using multipart API requests.

🚀 Deployment Options

🔹 Heroku

heroku create your-app-name
git push heroku main

🔹 Docker

docker build -t plant-disease-detection .
docker run -p 8000:8000 plant-disease-detection

🐛 Common Issues

Model not loading?

ls models/plant_disease_model.h5

Dependency issues?

pip install -r requirements.txt --force-reinstall

CORS error?

CORS_ALLOW_ALL_ORIGINS = True

🤝 Contributing

Contributions are welcome!

1. Fork the repo
2. Create a branch
3. Commit changes
4. Open PR

📝 License

MIT License

👨‍💻 Author

Somnath Das

• GitHub: @dassomnath99
• Email: somnathdas4462@gmail.com

🙏 Acknowledgments

• PlantVillage Dataset (Kaggle)
• TensorFlow Team
• Django Community

💡 Final Thoughts

This project is more than just a machine learning model — it’s a step toward smart agriculture 🌍

By combining AI + Web Development, we can empower farmers with tools that are:

• Fast
• Accurate
• Accessible

💬 If you found this useful, consider starring the repo and sharing your feedback!