This project presents an AI and IoT-based solution for detecting Yellow Leaf Disease in arecanut plantations and performing targeted pesticide spraying using a drone-mounted system. It leverages ResNet-50 for classification and an ESP32-CAM module for real-time monitoring.

• 🔍 Real-time disease detection using ResNet-50
• 🧠 Classifies leaves into Healthy, Yellow Leaf, Other, and No Leaf
• 🚁 Autonomous pesticide spraying via ESP32 and 12V pump
• 🧰 Drone-compatible, lightweight and modular hardware design
• 📈 Achieved 99.35% validation accuracy

• Python, TensorFlow, Keras, OpenCV
• ESP32-CAM, 12V DC Pump, Relay
• CREO for CAD design
• 3D printing and laser-cut hardware
• Drone platform for deployment

code/           → Jupyter notebook and model code
models/         → Trained ResNet50 model
hardware/       → Circuit diagrams, CAD models, drone images
video/          → Drone demonstration video
docs/           → Final report

Install the required libraries:

pip install -r requirements.txt

1. Image Capture – Drone captures leaf images via ESP32-CAM.
2. Classification – ResNet-50 model detects diseased leaves.
3. Trigger Relay – Activates 12V pump to spray pesticide.
4. Precision Spraying – Sprays for exactly 5 seconds.
5. Repeat – Continues as the drone flies across the plantation.

• Healthy Arecanut Leaf

• Arecanut Yellow Leaf Disease

• Other Leaf

🎥 Watch Drone in Action

See detailed methodology, hardware components, testing results, and modeling in the Final Project Report.

📰 This project is officially published in IEEE.
📖 Read it here: IEEE Xplore - Yellow Leaf Disease Detection and Autonomous Aerial Spraying

This project is licensed under the MIT License.

• Veeresha R.K.
• Shilpa M.K.
• Lathish Kumar N D
• Swaroop
• Samarth S Shetty
• Shrajan G Prasad

📢 Presented at IEEE ICRASET 2024
📄 Patent application submitted.