🌿 Leaf Disease Detection (CNN)

CNN-based plant disease classifier achieving 86.78% accuracy across 10+ disease categories — trained on the PlantVillage dataset using TensorFlow and transfer learning.

---

📌 Overview

This project uses a CNN to detect and classify plant leaf diseases from images. Early detection of plant diseases can dramatically reduce crop losses and improve agricultural yield.

---

✨ Key Features

• ✅ 86.78% accuracy on 10+ plant disease categories
• ✅ Transfer learning with pretrained CNN architectures
• ✅ Data augmentation to handle limited labeled data
• ✅ Interactive Streamlit web app for real-time prediction
• ✅ OpenCV preprocessing pipeline
• ✅ Trained on PlantVillage dataset

---

📁 Project Structure

Leaf-disease-prediction/
├── Leaf_Diseases_Prediction.ipynb   # Model training notebook
├── app.py                           # Streamlit web app
├── plant_disease_model.h5           # Trained model weights
├── requirements.txt
└── README.md

---

🚀 Getting Started

git clone https://github.com/MishraAbhay03/Leaf-disease-prediction.git
cd Leaf-disease-prediction
pip install -r requirements.txt
streamlit run app.py

---

📊 Model Performance

Metric	Score
Accuracy	86.78%
Dataset	PlantVillage
Classes	10+
Framework	TensorFlow/Keras

---

👤 Author

Abhaykumar Mishra — GitHub · LinkedIn

---

📄 License

MIT License — see LICENSE for details.