🌱 Crop Risk Detection and Yield Prediction

A deep learning–based system for crop disease detection, severity estimation, and yield loss prediction using plant leaf images. The project leverages computer vision and CNN models to help farmers and agricultural analysts identify diseases early and estimate potential crop loss.

---

🚀 Project Overview

Crop diseases significantly reduce agricultural productivity worldwide. Early detection of diseases and estimating their severity can help prevent large yield losses.

This project provides an AI-powered pipeline that:

• Detects crop diseases from leaf images
• Estimates disease severity level
• Predicts potential yield loss percentage
• Demonstrates applications in precision agriculture

---

🧠 Key Features

• 🌿 Crop Disease Detection using deep learning image classification
• 📊 Severity Estimation to measure disease impact on leaves
• 📉 Yield Loss Prediction based on detected disease severity
• 🧹 Image preprocessing and normalization pipeline
• ⚡ End-to-end prediction workflow from image → risk estimation

---

🏗 Project Pipeline

Leaf Image
     ↓
Image Preprocessing
     ↓
Disease Classification (CNN Model)
     ↓
Severity Estimation
     ↓
Yield Loss Prediction
     ↓
Risk Output

---

🛠 Tech Stack

Programming Language

• Python

Machine Learning / Deep Learning

• TensorFlow / Keras
• CNN Models
• Transfer Learning

Computer Vision

• OpenCV
• Image preprocessing

Tools

• Jupyter Notebook
• NumPy
• Matplotlib
• Scikit-learn

---

📂 Project Structure

Crop-Risk-Detection-and-Yield-Prediction
│
├── dataset/
│   └── plant leaf images
│
├── models/
│   └── trained CNN models
│
├── notebooks/
│   └── crop_risk_detection.ipynb
│
├── results/
│   └── model predictions and evaluation
│
├── requirements.txt
└── README.md

---

📊 Model Performance

Model	Task	Result
CNN Model	Disease Classification	High accuracy
Severity Estimation	Disease Impact Detection	Effective
Yield Prediction	Crop Loss Estimation	Demonstrated prediction capability

(Performance depends on dataset size and training configuration.)

---

💡 Applications

• Precision Agriculture
• Smart Farming Systems
• Crop Health Monitoring
• Early Disease Detection
• Agricultural Decision Support

---

🔮 Future Improvements

• Integrate segmentation models for better severity estimation
• Deploy as a web or mobile application for farmers
• Expand to support more crops and diseases
• Use larger agricultural datasets for improved accuracy

---

👨‍💻 Author

Harsha Vardhan Appikatla

• GitHub: https://github.com/HarshaAppikatla
• LinkedIn: https://www.linkedin.com/in/appikatlaharshavardhan/

---

📜 License

This project is licensed under the MIT License.