The BetterFarm Bird Detection System utilizes a Raspberry Pi, a camera, and sophisticated computer vision techniques to automatically detect birds in real-time. If a bird is detected, a message is sent to a farmer through Telegram, alerting them of the bird's presence.
Motion Detection: By comparing consecutive frames, we can identify any movement on the field, potentially indicating bird activity. Bird Identification with YOLOv3: Once motion is detected, the system uses the YOLOv3 algorithm to confirm if the movement was caused by a bird. Alerts via Telegram: When a bird is detected, a Telegram bot sends a notification message to the farmer.
Raspberry Pi (with Raspbian OS installed) Camera module for Raspberry Pi YOLOv3 weights (download link needed in bird_detect.py)
- Clone this repository.
- Install necessary Python packages:
pip install -r requirements.txt
- Ensure you have the YOLOv3 weights in the yolo_data folder.
- Set up the Telegram bot and get the TOKEN.
- Replace the TOKEN in the Flask server script.
python TelegramBot/main.py
python main.py
When the system detects a bird, it will save the image to a local directory, and you will receive an alert on Telegram.
Samah Rajabi, Rahid Abounis, Noam Mirjani