BotanIQ - Tomato Plant Disease Detection AI

An AI-powered mobile-friendly web application for identifying tomato plant diseases using computer vision and machine learning. Perfect for farmers, gardeners, and agricultural professionals.

🌟 Features

• Disease Detection: Identifies 5 tomato diseases with high accuracy
• Mobile Optimized: Works on phones, tablets, and desktops
• Offline Capable: Runs completely offline after initial setup
• Voice Feedback: Audio descriptions of diagnoses
• Real-time Analysis: Instant results from leaf photos
• User-friendly Interface: Simple upload and analyze workflow

📱 Mobile Usage

✅ Works on Mobile Phones

BotanIQ is designed to work seamlessly on mobile devices:

• Camera Integration: Take photos directly from your phone's camera
• Touch Optimized: Large buttons and mobile-friendly interface
• Responsive Design: Adapts to any screen size
• Fast Loading: Optimized for mobile networks and data usage

🔄 Offline Mobile Usage

Yes, BotanIQ works completely offline on mobile devices!

What Works Offline:

• ✅ Disease detection and analysis
• ✅ Voice feedback (on supported mobile browsers)
• ✅ All core functionality
• ✅ Saved model predictions

What Requires Internet (One-time):

• ⏳ Initial app loading and model download
• ⏳ Streamlit server connection (if running remotely)

📲 Mobile Deployment Options

Option 1: Local Network Access (Recommended)

# Run on your computer
streamlit run scripts/app.py

# Access from phone via local network:
# http://YOUR_COMPUTER_IP:8501

Option 2: Cloud Deployment

Deploy to services like:

• Streamlit Cloud (free tier available)
• Heroku
• AWS/GCP/Azure
• Railway or Render

Recommended cloud build/start commands

pip install --upgrade pip
pip install -r requirements.txt
streamlit run scripts/app.py --server.port $PORT --server.headless true --server.address 0.0.0.0

Option 3: Mobile App Conversion

Convert to native app using:

• PWA (Progressive Web App) capabilities
• Capacitor or Cordova for native apps
• React Native rebuild

🚀 Quick Start

Prerequisites

• Python 3.11+
• Windows/macOS/Linux
• 4GB+ RAM recommended

Installation

1. Clone/Download the project

   git clone <repository-url>
   cd botanq_ai

2. Create virtual environment

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

3. Install dependencies

   pip install -r requirements.txt

4. Configure HuggingFace token Create .streamlit/secrets.toml:

   HF_TOKEN = "your_huggingface_token_here"

   Get a token at: https://huggingface.co/settings/tokens

5. Run the app

   streamlit run scripts/app.py

   The app will open at http://localhost:8501

📱 Using on Mobile

Method 1: Local Network Access

1. Run the app on your computer as described above
2. Find your computer's IP address:
   • Windows: ipconfig in Command Prompt
   • macOS: ifconfig in Terminal
   • Linux: ip addr show
3. On your phone, open browser and go to: http://YOUR_IP:8501

Method 2: Direct Phone Access

If running on the same device:

• Use http://localhost:8501 on the device itself
• Or use http://127.0.0.1:8501

Method 3: Cloud Deployment

Deploy to a cloud service and access via public URL from any device.

🔊 Text-to-Speech (TTS)

Desktop (Windows)

• Uses Windows built-in System.Speech API
• No extra packages required
• Works completely offline
• Multiple voice options available

Mobile Browsers

• Uses browser's built-in speech synthesis
• May require user permission for audio
• Works offline after page loads
• Voice quality depends on device/browser

📊 Supported Diseases

1. Bacterial Spot - Small dark spots with yellow halos
2. Early Blight - Dark spots with concentric rings
3. Late Blight - Water-soaked lesions, white mold
4. Septoria Leaf Spot - Small circular spots with dark borders
5. Healthy - No disease symptoms

🛠️ Troubleshooting

Common Issues

App won't load on mobile:

• Check firewall settings
• Ensure both devices are on same network
• Try different browser (Chrome recommended)

TTS not working:

• On mobile: Check browser permissions for audio
• On desktop: Ensure Windows speech voices are installed when running locally
• On Render/Railway: cloud TTS may not be available; use pre-generated mp3 with st.audio() or handle gracefully in code

Model download fails:

• Check internet connection
• Verify HuggingFace token is correct
• Try again (may take time on slow connections)

Camera not working:

• Grant camera permissions in browser
• Try different browser
• Check device camera settings

Performance Tips

• Close other apps when running analysis
• Use clear, well-lit photos for best results
• Restart app if it becomes slow
• Clear browser cache if interface issues occur

📈 Model Information

• Framework: TensorFlow/Keras
• Architecture: Convolutional Neural Network (CNN)
• Input: 256x256 RGB images
• Output: 5-class classification
• Accuracy: ~95% on test dataset
• Size: ~50MB (downloads automatically)

🤝 Contributing

Contributions welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Tomato disease dataset from PlantVillage
• TensorFlow/Keras for ML framework
• Streamlit for web interface
• HuggingFace for model hosting

---

Made with ❤️ for farmers and gardeners worldwide