README.md

metadata

title: 'CropGuard: Leaf Disease Detector'
colorFrom: green
colorTo: indigo
sdk: gradio
sdk_version: 5.27.1
app_file: app.py
pinned: false

CropGuard: Leaf Disease Detector

CropGuard is a lightweight, deployable machine learning app that detects leaf diseases in Potato, Tomato, and Grape plants from user-uploaded or captured images.

Built using PyTorch, Gradio, Docker, and Hugging Face Spaces, it provides the following capabilities:

• Upload or capture a leaf image
• Predict plant health status
• Identify likely disease (if any)
• Visualize model attention using GradCAM++ heatmaps
• Provide quick disease information and treatment suggestions

---

Project Structure

CropGuard/
├── app.py                # Gradio app
├── Dockerfile            # Docker container definition
├── requirements.txt      # Python dependencies
├── src/                  # Source code (organized into modules)
│   ├── app/
│   ├── data/
│   ├── model/
│   └── utils/
├── sample_images/        # Few test images (optional for demo)
├── disease_info.json     # Disease descriptions
└── README.md             # (this file)

License

MIT License.

---

Acknowledgments

• Dataset: PlantVillage Dataset
• Base Model: MobileNetV2 (pretrained on ImageNet)
• Visualization: GradCAM++

---

Author

Made by Arka Mitra © 2025.

---