🚨 CrisisMind AI — Emergency Intelligence Platform

"Transform any emergency photo into an actionable rescue plan in under 10 seconds."

Repository • Submission Showcase • Demo App

---

🏆 Project Overview

CrisisMind AI is an emergency intelligence platform engineered using Google's AI and developer stack. When a citizen or field responder captures or uploads a photo of an accident, natural disaster, or fire, CrisisMind AI analyzes the scene in under 10 seconds to:

1. Detect Hazards & Objects: Identifies structural damage, fire/smoke risk, fluid leaks, crushed vehicles, and road blockades.
2. Estimate Casualty Triage: Counts occupants, identifies severe physical injuries, and recommends specialized trauma services.
3. Generate Executive AI Incident Summaries: Synthesizes complex scene context into concise, high-priority briefs.
4. Forecast Predictive Risk Progression: Computes a time-series escalation timeline (0–2m, 2–5m, 5–10m).
5. Connect Live Response Nodes: Locates nearby hospitals, police hubs, 108 ambulances, and relief shelters via Google Maps.
6. Provide Hands-Free Voice Guidance: Delivers real-time spoken instructions using Web Speech synthesis.
7. Broadcast Multilingual SOS Alerts: Translates emergency notifications into regional languages (Hindi, Marathi, Gujarati, Tamil, etc.).

---

👥 Development Team

• Aryan Meena — Lead Developer (aryan.meena@crisismind.ai)
• Lakshaya Kumawat — Collaborator (lakshaya.kumawat@crisismind.ai)

---

📷 Verified Emergency Demo Scenarios

Scenario 1: Multi-Vehicle Intersection Collision

• Input Scene: Red hatchback with severe side compression, white passenger sedan, downed motorcycle, crosswalk asphalt, and gathered bystanders at St. Wilfred College intersection.
• AI Analysis Output:
  • Severity: 🔴 HIGH / CRITICAL (97% Confidence)
  • Detected Vehicles: 3 Vehicles (Red Hatchback, White Sedan, Motorcycle)
  • Identified Hazards: Multi-Vehicle Crash, Intersection Traffic Blockade, Radiator & Fluid Leakage, Shattered Glass Debris.
  • Immediate Protocol: Divert oncoming intersection traffic, immobilize motorcyclist cervical spine, disconnect car battery cables, request 108 Trauma Ambulance.

---

Scenario 2: Severe Earthquake & Structural Collapse

• Input Scene: Residential neighborhood showing 45° tilted two-story buildings, collapsed tile roofs, downed utility lines, and debris blocking narrow access roads.
• AI Analysis Output:
  • Severity: 🔴 CRITICAL (98% Confidence)
  • Detected Objects: Tilted Two-Story House, Collapsed Tile Roof, Downed Utility Lines, Rubble Pile, Safety Cones.
  • Primary Hazards: Structural Collapse Hazard, Tilted Leaning Buildings, Downed Live Power Lines, Road Debris Blockade.
  • Immediate Protocol: 30-meter exclusion perimeter around leaning buildings, isolate local electrical grid, deploy Urban Search and Rescue (USAR) acoustic search probes.

---

🛠️ Tech Stack (Google-First Architecture)

Component	Tool / Technology	Purpose
Vision & Multimodal AI	Gemini 2.0 Flash (@google/genai)	Analyzes scene photos with structured JSON schema for instant triage.
Mapping & Navigation	Google Maps Embed API	Displays interactive maps, nearby response units, and GPS routing.
UI Design System	Material Design 3 + Glassmorphism	Modern Google dark theme (#0F172A) with Google brand accents (#4285F4, #EA4335, #FBBC05, #34A853).
Frontend Framework	React 18 + Vite 8 + TypeScript	High-performance SPA with instant HMR and strict type safety.
Styling & Animations	Tailwind CSS v4 + Framer Motion	Micro-interactions, radar sweeps, and responsive glassmorphic cards.
Voice Guidance	Web Speech Synthesis API	Spoken audio readouts for hands-free rescue operation.
Analytics Dashboard	Recharts	Telemetry charts showing incident distribution, risk levels, and response times.

---

✨ Key Features

⚡ 1. Live Device Camera Capture & Drag-and-Drop Scanner

• Real-time webcam integration via navigator.mediaDevices.getUserMedia allows first responders to capture snapshot photos directly from their phone or computer camera.
• Drag-and-drop file uploader supporting PNG, JPG, and JPEG up to 15MB.

🧠 2. Gemini Multimodal Scene Analysis

• Executive AI Incident Summary: Concise 2-sentence situational briefing powered by Gemini multimodal reasoning.
• Injury & Hazard Triage: Identifies cervical spine trauma, limb fractures, dust inhalation, fuel leaks, and fire risks.
• Occupant & Vehicle Detection: Quantifies people count and vehicle involvement.

⏱️ 3. Predictive AI Incident Timeline

• Time-series risk progression model forecasting threats:
  • 0 – 2 Min: Immediate threat containment & perimeter isolation.
  • 2 – 5 Min: Secondary escalation risks & traffic bottleneck prevention.
  • 5 – 10 Min: Medical triage window & rescue unit arrival stabilization.

🗺️ 4. Google Maps Live Response Nodes

• Interactive map embedding nearby emergency infrastructure:
  • 🏥 Level-1 Trauma Centers & Super Specialty Hospitals
  • 🚓 Central Control Police Hubs (112)
  • 🚑 108 Advanced Life Support Ambulances
  • ⛺ Civil Defense Relief Shelters

📢 5. Multilingual SOS Generator

• One-tap alert generator with single-click Copy, WhatsApp sharing, and SMS dispatch.
• Translates messages on the fly into Hindi, Marathi, Gujarati, Tamil, and English.

📊 6. Command Center Telemetry

• Real-time analytics dashboard displaying severity breakdowns, incident categories, and service dispatch metrics.

---

🚀 Quick Start Guide

Prerequisites

• Node.js (v18 or higher)
• npm or yarn

Installation Steps

1. Clone the Repository

   git clone https://github.com/Aryan-sourcee/Google_X_JECRC.git
   cd Google_X_JECRC

2. Install Dependencies

   npm install

3. Configure Environment Variables Create a .env file in the project root:

   VITE_GEMINI_API_KEY="your_gemini_api_key_here"
   VITE_GOOGLE_MAPS_API_KEY="your_google_maps_api_key_here"

4. Run Development Server

   npm run dev

   Open your browser at http://localhost:5173/.

5. Build for Production

   npm run build

---

📁 Project Structure

Google_X_JECRC/
├── src/
│   ├── assets/              # Emergency disaster & crash test images
│   │   ├── intersection-car-crash.png  # Multi-vehicle crash demo image
│   │   └── earthquake-disaster.png     # Structural collapse demo image
│   ├── components/          # React views & UI components
│   │   ├── Header.tsx       # Google-styled main menu bar & status pills
│   │   ├── BottomNav.tsx    # Mobile navigation bar
│   │   ├── SplashView.tsx   # Hero landing screen & team credits
│   │   ├── HomeDashboardView.tsx # Responder control dashboard
│   │   ├── UploadView.tsx   # Drag & drop uploader + webcam camera modal
│   │   ├── ScanningOverlayView.tsx # 360° radar sweep & progress gauge
│   │   ├── ResultsView.tsx  # Emergency analysis & AI incident summary
│   │   ├── IncidentTimeline.tsx # Time-series forecasting cards
│   │   ├── MapServicesView.tsx  # Google Maps Live Embed
│   │   ├── SOSGeneratorView.tsx # Multilingual SOS alert box
│   │   ├── VoiceAssistantView.tsx # Web Speech audio guidance
│   │   ├── HistoryView.tsx  # Searchable emergency log history
│   │   ├── AnalyticsDashboardView.tsx # Recharts telemetry dashboard
│   │   └── LoginModal.tsx   # Team member profile switcher
│   ├── data/                # Sample incident presets & nearby services
│   ├── services/            # Gemini API & Web Speech services
│   │   ├── gemini.ts        # Gemini 2.0 Flash integration & prompt schema
│   │   └── speech.ts        # Web Speech synthesis module
│   ├── types/               # TypeScript interfaces & types
│   ├── App.tsx              # Main routing & application state
│   ├── index.css            # Design tokens & Tailwind CSS v4 setup
│   └── main.tsx             # Application entry point
├── AI_STUDIO_SHOWCASE.md   # Google AI Studio submission package & system prompt
├── README.md                # Project documentation with scenario images
└── vite.config.ts           # Vite build configuration

---

📄 License & Acknowledgements

Built for the Google Hackathon & Ecosystem Showcase. Special thanks to Google AI Studio, the Gemini API team, and Google Maps Platform.

Developed by Aryan Meena and Lakshaya Kumawat.