Poor Man Security Program

This project implements a basic webcam using the YOLOv8 object detection model to detect and record videos of people in real time. The application saves daily recordings, and old recordings are automatically deleted after a specified number of days.

Features

• Real-time person detection using YOLOv8.
• Saves video recordings only when a person is detected.
• Automatically deletes recordings older than a configurable number of days.
• Displays live video feed with bounding boxes for detected persons.
• Includes date and time overlay on the video feed.

Requirements

• Python 3.7+
• Webcam or camera device.
• Pre-trained YOLOv8 model.

Installation

1. Clone the repository:

   git clone https://github.com/KodyPrograms/Poor-Man-Security-Program.git
   cd Poor-Man-Security-Program

2. Install dependencies:

   python install.py

3. Ensure the YOLOv8 model is downloaded to the project directory (yolov8n.pt by default).

Usage

Run the application:

python security_camera.py

Configuration

• MODEL_PATH: Path to the YOLOv8 model file.
• RECORDINGS_DIR: Directory to save video recordings.
• DELETE_OLDER_THAN_DAYS: Number of days to retain old recordings.
• frame_width and frame_height: Video resolution settings. 720p by default.

Controls

• Press q to quit the application.

License

This project utilizes other dependencies such as Ultralytics YOLO and require them to function.