Object Detection with Email Notification and Audio-Video Recording

This project uses YOLOv4 (You Only Look Once) real-time object detection algorithm to detect objects in a video feed. It records audio and video clips, and sends email notifications with the latest recorded clips as attachments when an object in motion is detected.

Table of Contents

• Object Detection with Email Notification and Audio-Video Recording
  • Table of Contents
  • Introduction
  • Requirements and Installation

Introduction

This project utilizes the YOLOv4 Tiny model for object detection in a video feed. When an object in motion is detected, it records a 5-second audio and video clip, sends an email notification with the latest clips attached, and stores object detection information in a local SQLite database.

Requirements and Installation

To run this project, you need Python 3.7 or higher and the following libraries:

1. OpenCV (cv2)
2. NumPy
3. smtplib
4. email (built-in)
5. sqlite3 (built-in)
6. os (built-in)
7. time (built-in)
8. PyAudio
9. wave (built-in)
10. threading (built-in)

Install the required libraries using pip:

pip install opencv-python opencv-python-headless numpy PyAudio