/safe_guard_llm

Primary LanguagePython

Stadium Security System

Overview

This project implements a comprehensive security decision-making system for a sports stadium using AI and real-time data analysis. The system collects data from various sources, processes it, and uses a large language model to make security-related decisions.

Features

  • Real-time video stream analysis for pedestrian and vehicle traffic
  • Temperature and smoke sensor data collection
  • Continuous data processing and aggregation
  • AI-powered decision-making for security strategies
  • Asynchronous data collection and processing

Project Structure

stadium_security_system/

│

├── main.py

├── config.py

├── data_collectors/

│ ├── init.py

│ ├── camera_stream.py

│ ├── sensor_data.py

│ └── data_processor.py

├── prompt_engineering/

│ ├── init.py

│ └── prompt_generator.py

├── decision_making/

│ ├── init.py

│ └── llm_decision_maker.py

└── utils/

│ └── data_structures.py

├── config.py

└── main.py

Installation

  1. Clone the repository:

    git clone https://github.com/your-username/stadium-security-system.git
cd stadium-security-system

  2. Install the required dependencies:

    pip install -r requirements.txt

  3. Configure the system by editing config.py with your specific settings.

Usage

Run the main script to start the security system: python3 main.py

The system will begin collecting data from cameras and sensors, processing the information, and making security decisions at regular intervals.

Configuration

Edit the config.py file to customize:

  • Camera stream URLs and locations
  • Sensor IDs and locations
  • AI model API settings
  • Data processing and decision-making intervals

Components

  • Data Collectors: Gather data from cameras and sensors
  • Data Processor: Cleans and aggregates collected data
  • Prompt Generator: Creates prompts for the AI model based on processed data
  • AI Decision Maker: Interfaces with a large language model to make security decisions

License

This project is licensed under the MIT License - see the LICENSE file for details.