OmniGuard: Intelligent Disaster Monitoring System

OmniGuard is a real-time, multi-source disaster monitoring platform designed to provide immediate situational awareness and automated safety guidance. It aggregates data from global seismic sources (such as USGS), processes it via an event-driven architecture, and visualizes critical information on an interactive dashboard.

---

Key Features

• Real-Time Ingestion: Continuously monitors USGS earthquake data with 30-second polling intervals.
• Proximity Filtering: Automatically identifies events within a configurable radius of the user's location.
• AI-Powered Guidance: Leverages Google Gemini AI to provide instant, context-aware safety protocols for significant events.
• AI Chat Assistant: Interactive interface for follow-up inquiries and personalized emergency advice.
• Interactive Dashboard: Features a live Leaflet.js map with dynamic markers, radius visualization, and real-time event feeds via WebSockets.
• Resilient Architecture: Utilizes Apache Kafka to ensure data reliability and decoupling between ingestion and processing layers.
• Data Persistence: Records all event data in PostgreSQL (PostGIS) for historical analysis and reporting.

---

System Architecture

OmniGuard utilizes a containerized microservices architecture to ensure scalability and reliability

graph TD
    subgraph "External World"
        USGS[USGS API]
    end

    subgraph "OmniGuard Core"
        P[Producer] -->|Ingest| K{Apache Kafka}
        K -->|Stream| C[Smart Consumer]
        K -->|Stream| B[FastAPI Backend]
        
        C -->|AI Analysis| Gemini[Google Gemini]
        C -->|Persist| DB[(PostgreSQL)]
        
        B -->|Query| DB
        B <-->|WebSocket| UI[Web Dashboard]
    end
    
    USGS --> P

Loading

---

Project Structure

omniguard/
├── backend/            # FastAPI Backend & WebSocket Manager
│   └── main.py
├── frontend/           # Web Dashboard (HTML/CSS/JS)
│   ├── index.html
│   ├── style.css
│   └── app.js
├── consumer.py         # Smart Processing Service (AI & DB)
├── producer.py         # Data Ingestion Service (USGS)
├── docker-compose.yml  # Container Orchestration
├── view_db.py          # Database Inspection Utility
└── implementation_details.md # Technical Specifications

---

Getting Started

Prerequisites

• Docker and Docker Compose must be installed on the host machine.

Installation and Deployment

1. Clone the Repository:

git clone https://github.com/RustyyES/omniguard.git
cd omniguard

2. Configuration: Execute the setup script to configure environment variables and API credentials:

./setup.sh

3. Launch System: Deploy the full stack using the following command:

docker compose up --build -d

4. Access the Interface: Navigate to the following address in a web browser: http://localhost:8000

---

Usage Guide

1. Dashboard Overview

• Map Interface: Visualizes the user's location and recent seismic events with categorized markers.
• Live Feed: Provides instantaneous updates as new data is processed.
• Alert System: High-priority events trigger an alert overlay containing AI-generated safety instructions.
• AI Consultation: The integrated chat allows users to request specific information, such as shelter locations or utility shut-off procedures.

2. Simulation Mode

To validate alert workflows without active seismic activity:

1. Access the Simulator panel on the sidebar.
2. Select "Simulate Event Near Me".
3. The system will initiate a full alert cycle, including AI analysis and WebSocket notification.

3. Database Inspection

To query raw data stored within the system:

docker compose exec backend python view_db.py

---

Technical Stack

• Language: Python 3.13
• Backend Framework: FastAPI
• Message Broker: Apache Kafka & Zookeeper
• Database: PostgreSQL 15 with PostGIS extension
• Frontend: HTML5, CSS3, JavaScript (ES6), Leaflet.js
• Artificial Intelligence: Google Gemini Generative AI
• Orchestration: Docker

---

License

This project was developed for academic purposes as a Graduation Project.

---