README.md

metadata

title: Plant Disease Detector
emoji: 🍃
colorFrom: green
colorTo: green
sdk: gradio
sdk_version: 5.34.2
app_file: app.py
pinned: false

🍃 Plant Disease Detector

This is a Gradio-based web app for detecting fruit and leaf diseases using an EfficientNetB0-based CNN model.

🧠 Model Overview

• Architecture: EfficientNetB0 (transfer learning)
• Input Size: 160×160 RGB
• Classes: 21 fruit/leaf disease types (mango, guava, lime, pomegranate)
• Framework: TensorFlow / Keras
• Performance: ~99.5% train accuracy, ~98.5% validation accuracy

🖼️ How to Use

• Upload a fruit/leaf image via the interface
• The model will return the predicted disease name and confidence %
• You can also try it using the provided example images

📂 Example Images Sample images are available in the examples/ folder:

• Phytopthora.jpg
• RedRust.jpg
• HealthyMangoLeaf.jpg
• LimeLeafSpotted.jpg

🛠️ Dependencies All dependencies are listed in requirements.txt:

• tensorflow
• gradio
• numpy
• pillow

🙋‍♂️ Author Aarzoo Singh
Research Intern, Machine Learning
B.Tech CSE, NIT Patna
🔗 Hugging Face Profile

---

🌐 Live App: Click here to run this app