Link for presentation- guardianaipresentation.vercel.app
GuardianAI delivers AI-powered proactive personal safety—your silent guardian and immutable proof. The system monitors both auditory and visual cues for signs of distress, leveraging AI models to detect threats in real time from audio files and visual streams. Blockchain technology ensures all evidence is securely stored and tamper-proof, providing legal admissibility and trust.
- Real-time Threat Detection: Detects distress signals such as screams, cries for help, and threatening gestures using advanced AI models.
- Audio File Processing: Converts incoming audio into vector arrays using pretrained speech models. Evaluates confidence scores and classifies the environment as safe or unsafe.
- Blockchain-Certified Evidence: Stores audio-visual evidence on the Polygon blockchain for immutability and legal admissibility.
- App: The app records the environment audio when activated and this audio file is sent to the model to detect safe/unsafe environment.
- Legally Admissible Proof: Ensures all captured data is tamper-proof for court and legal proceedings.
- Visual + Audio AI: Uses technologies like TensorFlow and OpenCV for multi-modal detection.
- Monitor environment for streams of audio and video.
- Use AI models to analyze data for distress cues.
- Convert audio samples to vector arrays and compute confidence threshold for classification.
- Classify scene as ‘safe’ or ‘unsafe’; trigger alert if unsafe.
- Store relevant evidence in blockchain for secure retrieval.
- AI/ML: Pretrained speech & vision models (TensorFlow, PyTorch, OpenCV)
- Blockchain: Polygon (for tamper-proof storage)
- Programming Languages: Python, React for frontend, Node.js for backend
- Database: PostgreSQL
- App: frontend using React Native, Node.js for backend
- Additional Tools: Hugging Face, Torchaudio, Transformers, Google Gemini
- Proactive crime intervention and personal safety
- Evidence collection for legal support
- GuardianAl provides 24/7 protection for students commuting to and from campus.
- Real-time monitoring for healthcare workers, security guards, and remote employees in high-risk environments.
- Seamless integration with ride share and dating apps for enhanced safety.
- Enhanced detection
- Native Apps
- Professional Response
- Smart Infrastructure