DeepNudeDetect - Comprehensive NSFW/Deepfake Detection & Protection System

A robust tool for detecting deepnudes, deepfakes, NSFW, and NSFL content with image protection capabilities.

Features

🔍 Detection Capabilities

• Deepnude Detection: Identifies AI-generated nude imagery
• Deepfake Detection: Detects face manipulation and synthetic media
• NSFW/NSFL Detection: Classifies inappropriate content
• Multi-Model Analysis: Combines multiple AI models for robust detection

🛡️ Protection Features

• Image Perturbation: Add adversarial noise to protect against deepfake generation
• Privacy Protection: Fawkes-style cloaking to prevent facial recognition abuse

📊 Input Methods

• File upload
• Image URL analysis
• Folder scanning
• Removable drive scanning

🔌 Integrations

• AIorNot API
• OpenAI Vision API
• Google Gemini Vision
• Custom trained models

Architecture

DeepNudeDetect/
├── backend/          # FastAPI backend with ML models
├── frontend/         # React web interface
├── chrome-extension/ # Browser extension
├── models/          # Pre-trained ML models
└── data/            # Upload and processing directories

Components

Backend (Python + FastAPI)

• REST API for image analysis
• Multiple detection models
• Image perturbation algorithms
• API integrations

Web Interface (React)

• Upload interface
• Folder/drive scanning
• Real-time analysis results
• Image protection tools

Chrome Extension

• Automatic image detection
• Real-time alerts
• Privacy protection warnings

Installation

Backend Setup

cd backend
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
python download_models.py

Frontend Setup

cd frontend
npm install
npm run dev

Chrome Extension

1. Open Chrome and go to chrome://extensions/
2. Enable "Developer mode"
3. Click "Load unpacked"
4. Select the chrome-extension directory

Usage

Starting the Backend

cd backend
python main.py

Backend runs on http://localhost:8000

Starting the Web Interface

cd frontend
npm run dev

Frontend runs on http://localhost:3000

API Endpoints

• POST /api/analyze - Analyze single image
• POST /api/analyze-url - Analyze image from URL
• POST /api/analyze-folder - Scan entire folder
• POST /api/protect-image - Add adversarial perturbation
• GET /api/health - Health check

Configuration

Create .env file in backend directory:

OPENAI_API_KEY=your_key_here
GOOGLE_API_KEY=your_key_here
AIORNOT_API_KEY=your_key_here

Models Used

1. NudeNet - NSFW detection
2. DeepFake Detection Model - Face manipulation detection
3. CLIP - Content understanding
4. Custom CNN - Deepnude specific detection
5. Adversarial Perturbation - Fawkes-style protection

Security & Privacy

• All processing happens locally by default
• API keys are optional for enhanced detection
• Images are not stored unless explicitly saved
• Privacy-first approach

License

MIT License - Use responsibly for legitimate safety and security purposes only.

Disclaimer

This tool is designed for legitimate security research, content moderation, and personal protection. Misuse for harassment or malicious purposes is strictly prohibited.