This project implements an intelligent doorbell system utilizing YOLOv5 for object detection. It categorizes visitors and packages between distinct classes (e.g., person, Walmart delivery, parcel, mail) and alerts the homeowner via email or phone.
-
Real-time object detection using YOLOv5
-
Customizable class recognition
-
Efficient server-client architecture
-
Flexible email/phone notifications
-
Open-source framework (YOLOv5, Flask)
- Client:
- Captures video frames (OpenCV)
- Detects motion changes
- Sends frames to server
- Server:
- Receives frames
- Performs object detection (YOLOv5)
- Classifies detected objects
- Sends bounding boxes/labels back
- Client:
- Receives detection results
- Triggers email/phone alerts
- Python 3.x
- OpenCV
- Flask
- YOLOv5 (https://github.com/ultralytics/yolov5) Ensure you have the required dependencies installed by running
pip install -r requirements.txt
git clone https://github.com/SanketMagodia/A.I-Doorbell.git
pip install -r requirements.txt
in ./API
python app.py
in ./client
python app.py
Consider cloud storage, user authentication, and more robust notification mechanisms. Ensure compliance with privacy and security regulations.