LuminaCheck AI is an advanced AI-powered web application that detects whether a digital image is REAL, FAKE, or AI-GENERATED using Google Gemini Vision AI.
https://luminacheck-ai.streamlit.app
| Feature | Description |
|---|---|
| AI Detection | Detects REAL, FAKE, or AI-GENERATED images |
| Confidence Score | 0-100% confidence percentage |
| Detailed Reasoning | 2-3 specific visual clues |
| Detection History | Session-based history table |
| CSV Download | Download detection report |
| AI Chatbot | Sidebar AI assistant |
| Dark UI | Professional animated interface |
| Responsive | Works on mobile & desktop |
- Python — Core programming language
- Streamlit — Web application framework
- Google Gemini AI — Image analysis engine (gemini-2.5-flash)
- Pillow — Image processing
- Pandas — Data management
- Streamlit Cloud — Deployment & hosting
# Clone the repository
git clone https://github.com/codesbydevapriya/LuminaCheck-AI.git
# Install dependencies
pip install -r requirements.txt
# Set your API key
export GEMINI_API_KEY="your_api_key_here"
# Run the app
streamlit run app.pyLuminaCheck-AI/
├── app.py # Main application
├── requirements.txt # Dependencies
└── README.md # Project documentation
| Variable | Description |
|---|---|
GEMINI_API_KEY |
Google Gemini API Key |
Get your free API key at aistudio.google.com
- Upload any JPG/JPEG/PNG image
- Click "Analyze Image Now"
- AI scans for forensic clues
- Get verdict — REAL or FAKE
- Ultra-realistic AI images may occasionally be misclassified
- Free tier API has rate limits
- Detection accuracy depends on Gemini AI model
- Custom CNN model for higher accuracy
- Video deepfake detection
- Batch image processing
- Mobile application
Devapriya — BCA Final Year Student | March 2026
This project is licensed under the MIT License.
Made by Devapriya | Powered by Google Gemini AI