Enhancing Independence Computer Vision-Based Object Detection Techniques for the Visually Impaired
Priyanka More, Sachin Rambhau Sakhare, Rahul Shelke, Saurabh Raut, Yugandhar Patil, Darshan Vora · 2025
This paper introduces an innovative solution leveraging the advanced YOLOv8 deep learning model to provide real-time object detection for over 2 billion blind and visually impaired individuals. Acting as a virtual “eye,” the system allows users to recognize objects and environments with high precision and speed. By integrating the Google Text-to-Speech API, it delivers intuitive voice guidance, offering immediate audio feedback for identified objects. This approach addresses the key challenges of accessibility and independence faced by the visually impaired and opens doors for more advanced assistive technologies. The system combines YOLOv8's robustness, the versatility of Raspberry Pi, and the efficiency of text-to-speech to create a comprehensive object detection solution, designed to enhance users' navigation and boost their confidence in daily activities. This research represents a major step forward in assistive technology, offering practical solutions that significantly improve the quality of life for those with visual impairments.