AI Doorbell System

Description

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.

Demo

Key Features

• Real-time object detection using YOLOv5

• Customizable class recognition

• Efficient server-client architecture

• Flexible email/phone notifications

• Open-source framework (YOLOv5, Flask)

System Architecture

1. Client:
   • Captures video frames (OpenCV)
   • Detects motion changes
   • Sends frames to server
2. Server:
   • Receives frames
   • Performs object detection (YOLOv5)
   • Classifies detected objects
   • Sends bounding boxes/labels back
3. Client:
   • Receives detection results
   • Triggers email/phone alerts

Prerequisites

• 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

Usage/Examples

1. Clone the repository

git clone https://github.com/SanketMagodia/A.I-Doorbell.git

2. install dependencies

pip install -r requirements.txt

3. Run the server

in ./API

python app.py

3. Run the client on derired edge device

in ./client

python app.py

Notes

Consider cloud storage, user authentication, and more robust notification mechanisms. Ensure compliance with privacy and security regulations.

Contributions

Contributions are welcome! If you have any suggestions, feature requests, or improvements, feel free to open an issue or submit a pull request.