👁️ CrowdGuard

Panoptic Surveillance & Intelligent Crisis Response System

"Turning Chaos into Order." An AI-powered command center that synthesizes unstructured real-time data into actionable intelligence for emergency response.

🚀 Overview

CrowdGuard is a sophisticated crowd management and situational awareness platform designed for large venues. It utilizes the internal Aegis AI Engine (powered by Google Gemini) to analyze real-time surveillance feeds, detect anomalies, and coordinate emergency response units via a live, interactive dashboard.

💡 The Problem

In critical moments following a disaster, control rooms face:

• Information Overload: Thousands of panic calls and social posts.
• Unstructured Data: Hard-to-map images, voice recordings, and text.
• Slow Response: Manual verification delays critical dispatch decisions.

⚡ The Solution

CrowdGuard acts as a "Central Nervous System," instantly analyzing incoming media to extract geolocation and severity, visualizing threats on a heatmap, and automating unit dispatch.

---

✨ Key Features

1. 🗺️ Live Command Dashboard

• Real-time Map: Interactive SVG-based stadium map utilizing Leaflet and WebSockets for sub-second updates.
• Unit Tracking: Live GPS-style tracking of stewards, medics, drones, and police units.
• Heatmap Overlay: Thermography mode to visualize crowd density and risk levels dynamically.

2. 🤖 Aegis AI Analysis (Powered by Gemini)

• Forensic Breakdown: Analyzes CCTV images to identify specific threats (e.g., "Smoke plume," "Crushing").
• Risk Scoring: Automatically rates incidents (Safe, Moderate, Critical) and calculates density scores (1-10).
• Strategic Recommendations: AI generates actionable containment strategies (e.g., "Close Gate NE," "Deploy Medic to Section B").

3. 📢 Incident Management & Dispatch

• One-Click Deploy: Assign the nearest available unit to an incident directly from the map.
• Automated Reporting: Generate comprehensive executive situation reports summarizing all active zones.
• Public Advisory: AI drafts context-aware emergency alerts for public broadcast.

4. 🧪 Simulation Tools

• Feed Injector: A dedicated interface to simulate crowd reports, inject mock CCTV imagery, and test system responsiveness without real-world chaos.

---

🛠️ Tech Stack

Frontend (Client)

• Framework: React 18 + Vite
• Styling: Tailwind CSS + Framer Motion (Animations)
• Mapping: Leaflet Maps (React-Leaflet) + Custom SVG Overlays
• Icons: Lucide React

Backend (Server)

• Runtime: Node.js
• Framework: Express.js
• Database: MongoDB (Atlas) via Mongoose
• Real-Time: Socket.io (Bi-directional communication)

Artificial Intelligence

• Model: Google Gemini 2.0 Flash (via @google/genai SDK)
• Engine: Aegis Crowd Analysis AI

---

⚙️ Installation & Setup

Prerequisites

• Node.js (v18 or higher)
• MongoDB Atlas Account (Connection String)
• Google Gemini API Key (Get one at Google AI Studio)

1. Clone the Repository

git clone https://github.com/your-username/crowdguard.git
cd crowdguard

2. Backend Setup

Navigate to the server directory and install dependencies:

cd server
npm install

Create a .env file in the server folder with the following variables:

PORT=3001
MONGO_URI=mongodb+srv://<username>:<password>@cluster.mongodb.net/aegis_db
GEMINI_API_KEY=your_google_gemini_api_key

Start the backend server:

npm run dev
# Server will run on http://localhost:3001

3. Frontend Setup

Open a new terminal, navigate to the client directory, and install dependencies:

cd client
npm install

Create a .env file in the client folder (optional if running locally on default ports):

VITE_API_URL=http://localhost:3001

Start the frontend development server:

npm run dev
# Client will run on http://localhost:5173

---

📖 Usage Guide

1. Launch the Dashboard: Open http://localhost:5173/dashboard.
2. Launch the Simulator: Open http://localhost:5173/inject in a separate tab.
3. Simulate an Incident:

• Use the Feed Injector to upload an image of a crowd.
• Add a text description (e.g., "People pushing near North Gate").
• Click "Analyze & Inject".

4. Monitor & Respond:

• Watch the Dashboard as the AI processes the data.
• A red alert marker will appear on the map.
• Click the marker to view AI analysis.
• Click a nearby "Idle" unit to Deploy them to the scene.

---

📡 API Endpoints

The backend exposes the following RESTful endpoints:

Method	Endpoint	Description
POST	/api/analyze	Process image/text context via Gemini AI
POST	/api/generate-report	Generate a venue-wide situation report
GET	/api/incidents	Fetch all active incidents
POST	/api/incidents	Manually create an incident
GET	/api/units	Fetch all response units
POST	/api/units/deploy	Update unit status and location
POST	/api/reset	Clear all incidents and reset units

---

🤝 Contributing

1. Fork the repository.
2. Create a new branch (git checkout -b feature/AmazingFeature).
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a Pull Request.

---

📄 License

Distributed under the MIT License. See LICENSE for more information.