VigiByte - Global Criminal Intelligence Platform 🌍🔍

A multi-tenant AI-powered surveillance and criminal database system with real-time threat detection and global intelligence sharing.

🎯 Project Overview

VigiByte is an enterprise-grade security platform that enables organizations to:

• Monitor multiple surveillance cameras with AI-powered facial recognition
• Detect wanted criminals in real-time camera feeds
• Share global criminal database across all organizations
• Manage officers and viewers with role-based access control
• Track security alerts and maintain audit trails

✨ Key Features

🌐 Global Criminal Database

• Shared across all organizations - worldwide threat intelligence
• Read-only protection - immutable global records prevent tampering
• Persistent - survives organization deletion
• Instantly accessible - all users see updates in real-time

📹 Real-Time Surveillance

• Multi-camera support - manage unlimited surveillance nodes
• Live AI detection - facial recognition powered by TensorFlow.js
• Instant alerts - immediate notifications when criminals detected
• Performance monitoring - track CPU load, detection confidence, signal metrics

👥 Role-Based Access Control (RBAC)

• Admin - Full control, can approve/revoke officers, manage org data
• Officer - Can add cameras, manage local criminal database
• Viewer - Read-only access, view alerts and criminal records

🔐 Security Features

• 2FA Authentication - Google Authenticator support for admins
• Row-Level Security - Data isolation per organization
• Audit Logging - Complete activity trail
• Password Hashing - PBKDF2 encryption
• Session Management - JWT-based authentication

📊 Admin Dashboard Features

• Officer Management - Approve/revoke pending officers
• Real-time Subscriptions - Automatic updates via Supabase Realtime
• Organization Profiles - Multi-tenant data management
• System Health Monitoring - Connection status tracking

🏗️ Architecture

Tech Stack

• Frontend: React + Vite + Tailwind CSS
• Backend: Supabase (PostgreSQL)
• Authentication: Custom JWT + Supabase Auth
• AI/ML: TensorFlow.js (face-api.js)
• Real-time: Supabase Realtime
• Deployment: Vercel

Database Schema

organizations
├── id, domain, created_at

users
├── id, email, password_hash, role, organization_id
├── is_active, totp_secret, created_at

cameras
├── id, name, location, coordinates, type (webcam/ip/image)
├── source, organization_id, created_at

criminals
├── id, name, age, crime, danger_level
├── photo_url, organization_id, created_at

global_criminals ⭐ (READ-ONLY, NOT organization-scoped)
├── id, name, photo_url, is_wanted, threat_level
├── is_active, created_at

approved_officers
├── id, user_id, organization_id, created_at

🚀 Getting Started

Prerequisites

• Node.js 16+
• npm or yarn
• Supabase account
• Google Authenticator (for 2FA)

Environment Setup

1. Clone the repository

git clone https://github.com/FLASH-4/VigiByte.git
cd VigiByte

2. Create .env.local

cp .env.example .env.local
# Edit with your Supabase credentials

3. Install dependencies

npm install

4. Set up Supabase

• Create a Supabase project
• Run migrations (see Database Setup)
• Update VITE_SUPABASE_URL and VITE_SUPABASE_ANON_KEY

5. Start development server

npm run dev

🗄️ Database Setup

Create Tables in Supabase SQL Editor

-- Organizations
CREATE TABLE organizations (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  domain TEXT NOT NULL UNIQUE,
  created_at TIMESTAMP DEFAULT NOW()
);

-- Users
CREATE TABLE users (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  email TEXT NOT NULL UNIQUE,
  password_hash TEXT NOT NULL,
  role TEXT CHECK (role IN ('admin', 'officer', 'viewer')),
  organization_id UUID REFERENCES organizations(id),
  is_active BOOLEAN DEFAULT true,
  totp_secret TEXT,
  created_at TIMESTAMP DEFAULT NOW()
);

-- Cameras
CREATE TABLE cameras (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  name TEXT NOT NULL,
  location TEXT,
  coordinates TEXT,
  type TEXT CHECK (type IN ('webcam', 'ip', 'image')),
  source TEXT,
  organization_id UUID REFERENCES organizations(id),
  created_at TIMESTAMP DEFAULT NOW()
);

-- Organization Criminals (local database)
CREATE TABLE criminals (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  name TEXT NOT NULL,
  age INTEGER,
  crime TEXT,
  danger_level TEXT DEFAULT 'MEDIUM',
  photo_url TEXT,
  photo_data BYTEA,
  face_descriptor JSONB,
  organization_id UUID REFERENCES organizations(id),
  created_at TIMESTAMP DEFAULT NOW()
);

-- Global Criminals (SHARED, READ-ONLY)
CREATE TABLE global_criminals (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  name TEXT NOT NULL,
  photo_url TEXT,
  photo_data BYTEA,
  is_wanted BOOLEAN DEFAULT true,
  threat_level TEXT DEFAULT 'medium',
  description TEXT,
  is_active BOOLEAN DEFAULT true,
  created_at TIMESTAMP DEFAULT NOW()
);

-- Approved Officers
CREATE TABLE approved_officers (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  user_id UUID REFERENCES users(id),
  organization_id UUID REFERENCES organizations(id),
  created_at TIMESTAMP DEFAULT NOW()
);

-- Activity Log
CREATE TABLE audit_log (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  user_id UUID REFERENCES users(id),
  action TEXT NOT NULL,
  details JSONB,
  ip_address TEXT,
  created_at TIMESTAMP DEFAULT NOW()
);

Enable Row Level Security & Global Criminals Policy

ALTER TABLE users ENABLE ROW LEVEL SECURITY;
ALTER TABLE cameras ENABLE ROW LEVEL SECURITY;
ALTER TABLE criminals ENABLE ROW LEVEL SECURITY;
ALTER TABLE global_criminals ENABLE ROW LEVEL SECURITY;
ALTER TABLE approved_officers ENABLE ROW LEVEL SECURITY;

-- Global Criminals - READ-ONLY PROTECTION
CREATE POLICY "Global criminals read-only" ON global_criminals
  FOR SELECT USING (is_active = true);

CREATE POLICY "Prevent insert to global" ON global_criminals
  FOR INSERT WITH CHECK (false);

CREATE POLICY "Prevent update to global" ON global_criminals
  FOR UPDATE USING (false);

CREATE POLICY "Prevent delete from global" ON global_criminals
  FOR DELETE USING (false);

👤 User Workflows

Admin Registration

1. First admin creates account (org created automatically)
2. Sets up 2FA with Google Authenticator
3. Starts adding surveillance cameras
4. Approves officer registrations

Officer Registration

1. Registers with organization domain
2. Waits for admin approval
3. Once approved, can add local criminals
4. Can view global and local criminal databases

Viewer Access

1. Registered by admin as viewer
2. No editing permissions
3. Can view alerts and criminal records

🔄 Account Deletion Behavior

Admin Deletes Account

• ✕ Entire organization deleted
• ✕ All officers & viewers removed
• ✕ Organization cameras deleted
• ✕ Local criminal records deleted
• ✓ Global criminals preserved

Officer/Viewer Deletes Account

• ✓ Only user record deleted
• ✓ All organization data preserved
• ✓ Criminal records remain intact

🔒 Security Protocols

Password Requirements

• Minimum 8 characters
• At least 1 uppercase letter
• At least 1 lowercase letter
• At least 1 number
• At least 1 special character (@$!%*?&)

2FA Setup

• Google Authenticator app required
• QR code generation for easy setup
• Manual key entry for mobile users
• 6-digit TOTP verification

Rate Limiting

• Login attempts: 5 per minute per email
• Global throttling on sensitive endpoints

🚢 Deployment

Deploy to Vercel (Recommended)

# Connect to Vercel
vercel link

# Deploy
vercel

Docker Deployment

# Build image
docker build -t vigibyte:latest .

# Run container
docker run -p 3000:5173 vigibyte:latest

📱 Mobile Responsiveness

VigiByte is fully responsive with support for:

• Desktop (1920px+)
• Tablet (768px - 1919px)
• Mobile (320px - 767px)

🔧 Development

Available Scripts

npm run dev      # Start dev server
npm run build    # Build for production
npm run preview  # Preview production build

Folder Structure

src/
├── components/        # React components
│   ├── Dashboard.jsx
│   ├── AuthPanel.jsx
│   ├── CriminalDB.jsx
│   └── ...
├── lib/              # Utilities & services
│   ├── supabase.js
│   ├── faceRecognition.js
│   ├── organization.js
│   └── totp.js
├── services/         # Business logic
│   └── browserAuth.js
└── App.jsx

🐛 Troubleshooting

Storage Issues (Mobile)

On iOS Safari in private mode, localStorage may be unavailable. The app detects this and shows an appropriate warning.

Face Detection Not Working

• Check browser permissions
• Ensure camera feed is accessible
• Verify TensorFlow.js models are loaded
• Check browser console for errors

Real-time Updates Not Working

• Verify Realtime is enabled in Supabase
• Check RLS policies are correctly configured
• Ensure user has appropriate permissions

📊 Performance

• Face Detection: ~300-500ms per image (depends on CPU)
• Database Queries: <100ms average
• Real-time Updates: <1s propagation
• Build Size: ~850KB (gzipped)

📝 License

This project is proprietary. All rights reserved.

🤝 Contributing

Contributions are welcome! Please follow these guidelines:

1. Create a feature branch
2. Make your changes
3. Write clear commit messages
4. Submit a pull request

📧 Contact

For issues, questions, or feature requests, please create an issue on GitHub.

---

VigiByte - Making the world safer, one detection at a time. 🌍🔒