Leveraging AI and Vision Technologies for Safer Environments
To develop accessible thermal night vision tools that improve visibility and safety for at-risk communities during low-light conditions.
Improved detection, identity verification, and proactive alerts for community protection.
Uses YOLOv8 for human recognition and segmentation in real-time.
Applies thermal color mapping to human regions to highlight presence in darkness.
Matches real-time facial features against a known database to identify individuals.
Visual labels and alert sounds distinguish known vs unknown individuals.
Automatically brightens dark frames to ensure clearer visuals for processing.
Human shapes highlighted with inferno colormap for sharp contrast in visuals.
Alerts sound when humans are detected but not recognized.
Cross-platform audio notifications to draw immediate attention.
Enables safer environments in underserved regions, especially at night.
Potential for integration in community centers, transit hubs, and public areas.