Plant-Disease-Classifier

🌿 Plant Disease Classifier

AI-Powered Plant Disease Detection System
A production-ready deep learning web application that diagnoses diseases in plant leaves using advanced convolutional neural networks (CNNs).

Plant Disease Classifier Demo

🎥 Video Demonstration

✨ Features

🔬 Instant Disease Detection: Upload an image of a plant leaf and get immediate AI-powered diagnosis results
📊 Comprehensive Analysis: Provides detailed information on detected diseases, including causes, treatments, and prevention strategies
🎨 Professional Interface: Clean, intuitive UI with modern styling, image previews, and interactive visualization
🌱 Multi-Plant Support: Currently supports tomatoes, potatoes, and bell peppers with 15+ disease classifications
⚡ High Accuracy: Achieves 96.5% accuracy on test datasets with production-grade performance
📱 Responsive Design: Works seamlessly across desktop and mobile devices
🖼️ Example Gallery: Try the application instantly with pre-loaded example images

🧪 Supported Plant Diseases

The model can accurately identify the following plants and diseases:

🍅 Tomato

✅ Healthy
🦠 Bacterial Spot
🍂 Early Blight
🍂 Late Blight
🍄 Leaf Mold
🍃 Septoria Leaf Spot
🕷️ Spider Mites
🎯 Target Spot
💛 Yellow Leaf Curl Virus
🦠 Mosaic Virus

🥔 Potato

✅ Healthy
🍂 Early Blight
🍂 Late Blight

🌶️ Bell Pepper

✅ Healthy
🦠 Bacterial Spot

🔧 Model Architecture

The application uses a sophisticated custom CNN architecture with the following components:

5 Convolutional Blocks: Advanced feature extraction with batch normalization and ReLU activation
Global Average Pooling: Reduces overfitting and improves generalization
Fully Connected Layers: Dense layers with dropout regularization for robust classification
Data Augmentation: Extensive preprocessing pipeline with rotation, scaling, and color adjustments
Transfer Learning: Trained on the comprehensive PlantVillage dataset
Production Optimization: ONNX export for deployment efficiency

📈 Performance Metrics

Test Accuracy: 96.5%
Training Dataset: 54,000+ images
Validation Split: Stratified 70/15/15 split
Model Size: Optimized for production deployment
Inference Time: < 2 seconds per image

🛠️ Technology Stack

Backend & AI

PyTorch 2.0+: Deep learning framework
torchvision: Computer vision utilities
scikit-learn: Data preprocessing and evaluation
NumPy & Pandas: Data manipulation
PIL/Pillow: Image processing

Frontend & Deployment

Streamlit: Interactive web application framework
Matplotlib & Plotly: Data visualization
CSS3: Custom styling and responsive design
HTML5: Semantic markup

Development & Production

Python 3.8+: Core programming language
Git: Version control
Streamlit Cloud: Deployment platform
ONNX: Model optimization and export

🚀 Quick Start

Prerequisites

Python 3.8 or higher
pip package manager
4GB+ RAM recommended

Installation

Clone the repository

git clone https://github.com/Subrata3841/Plant-Disease-Classifier.git
cd Plant-Disease-Classifier

Install dependencies
```
pip install -r requirements.txt
```
Run the application
```
streamlit run app.py
```
Access the application
- Open your browser and navigate to http://localhost:8501
- Upload a plant leaf image or try the example images
- Get instant AI-powered disease diagnosis

📁 Project Structure

Plant-Disease-Classifier/
├── app.py                          # Main Streamlit application
├── plant_disease_classifier.py     # Core ML model and training logic
├── notebook.ipynb                  # Jupyter notebook for experimentation
├── requirements.txt                # Python dependencies
├── README.md                       # Project documentation
├── models/                         # Trained model files
│   ├── best_model.pth             # PyTorch model weights
│   ├── class_names.json           # Disease class mappings
│   ├── inference_transform.pkl    # Image preprocessing pipeline
│   ├── label_encoder.pkl          # Label encoding utilities
│   ├── model_config.json          # Model configuration
│   └── plant_disease_model.onnx   # ONNX optimized model
└── images/                         # Assets and examples
    ├── logo.png                   # Application logo
    └── examples/                  # Sample test images
        ├── tomato_healthy.jpg
        ├── Potato_Late_blight.jpeg
        └── Pepper_bell_Bacterial_spot.jpeg

👥 Development Team

Devansh Sengar
ML Engineer & Backend Developer

Udit Jain
AI Researcher & Model Architect

Subrata Mondal
ML Engineer & Backend Developer

📜 License

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

📞 Support

For questions, issues, or suggestions:

🐛 Issues: GitHub Issues
💬 Discussions: GitHub Discussions

🙏 Acknowledgments

PlantVillage Dataset: For providing the comprehensive plant disease dataset
PyTorch Community: For the excellent deep learning framework
Streamlit Team: For the intuitive web application framework
Open Source Community: For the amazing tools and libraries that made this project possible