README.md

metadata

title: Rice Disease Detection API v3
emoji: 🌾
colorFrom: green
colorTo: yellow
sdk: docker
pinned: false
license: mit
app_port: 7860

🌾 Rice Disease Detection API v3

EfficientNet-B0 based rice leaf disease detection API

Features

• 🔬 Detects 14 rice diseases + healthy plants
• 🎯 ~89-93% accuracy using EfficientNet-B0
• ✅ Rice leaf validation before disease detection
• 🚀 Fast inference with FastAPI

API Endpoints

Endpoint	Method	Description
/	GET	API info
/health	GET	Health check
/status	GET	Model status
/validate-leaf/	POST	Check if image is rice leaf
/predict/	POST	Predict disease from image

Usage

Predict Disease

curl -X POST "https://YOUR-SPACE.hf.space/predict/" \
  -H "accept: application/json" \
  -F "file=@rice_leaf_image.jpg"

Response Example

{
  "is_rice_leaf": true,
  "disease": "rice_blast",
  "confidence": "92.45%",
  "recommendation": "Use resistant varieties...",
  "category": "Fungal",
  "severity": "Very High"
}

Diseases Detected

1. Bacterial Leaf Blight
2. Bacterial Leaf Streak
3. Bakanae
4. Brown Spot
5. Grassy Stunt Virus
6. Healthy Rice Plant
7. Narrow Brown Spot
8. Ragged Stunt Virus
9. Rice Blast
10. Rice False Smut
11. Sheath Blight
12. Sheath Rot
13. Stem Rot
14. Tungro Virus

Model Details

• Architecture: EfficientNet-B0 (transfer learning from ImageNet)
• Training: Custom dataset with 17,789 images
• Framework: PyTorch 2.x

Project

Part of the SUMVILTAD Agricultural Management System.

License

MIT License