A Vision-Based Wayfinding System for Visually Impaired People Using Situation Awareness and Activity-Based...

Area of Science:

• Computer Science
• Human-Computer Interaction
• Assistive Technology

Background:

• Wayfinding presents a significant challenge for visually impaired individuals navigating unfamiliar environments.
• Existing assistive technologies often lack real-time situational awareness and object recognition capabilities.

Purpose of the Study:

• To develop and evaluate a novel smartphone-based wayfinding system for the visually impaired.
• To enable real-time situation and scene object recognition for enhanced navigation.

Main Methods:

• The system analyzes streaming images for real-time situation classification and context object interpretation using computer vision.
• Inertial sensors estimate user motion and record trajectories for route guidance and return trips.
• Activity-based auditory instructions are generated for intuitive user guidance.

Main Results:

• Situation awareness accuracy reached 90% with an object detection false alarm rate of 0.016 in indoor experiments.
• Field tests demonstrated a 97% accuracy in users locating their paths using the system.

Conclusions:

• The developed wayfinding system effectively assists visually impaired individuals in navigating indoor environments.
• Real-time scene analysis and auditory feedback significantly improve wayfinding performance and user confidence.