NJ Beach Cam Object Detection Notifier

This project downloads a specific time range from a live stream of a NJ beach, detects objects (cars and people) in using OpenCV, and sends Twilio notifications when a threshold of detected cars is met.

Features

• Download specific portions of YouTube videos.
• Detect cars and people in a video stream.
• Send Twilio notifications when a car detection threshold is reached.
• Output video stream to a web browser via Flask.

Setup

Prerequisites

• Python 3.x
• yt-dlp for downloading videos.
• OpenCV for video processing.
• Twilio for notifications.

Installation

1. Clone the repository and navigate to the project directory.

   git clone https://github.com/yourusername/yt-dlp-object-detection.git
   cd yt-dlp-object-detection
   

2. Install dependencies

   pip install -r requirements.txt
   

3. Set up Twilio by adding your account details to environment variables.

   export TWILIO_ACCOUNT_SID='your_account_sid'
   export TWILIO_AUTH_TOKEN='your_auth_token'
   export TWILIO_PHONE_NUMBER='your_twilio_number'
   export YOUR_PHONE_NUMBER='your_phone_number'
   

4. Add the YouTube video URL in the configuration file.

   YOUTUBE_URL = "https://www.youtube.com/watch?v=your_video_id"
   

Run Project

The Flask app processes the video stream, detects objects (cars and people), and sends notifications when a threshold of detected cars is reached.

1. Run the Flask app to initiate the video stream and object detection.

   python app.py
   

2. Visit http://localhost:5000 to view real-time detection results.