Obstacle Avoidance for blind people using Yolo algorithm, Darknet and GTTS

Rajalakshmi Institute of Technology, Chennai, India

^* shriram.cs.2019.cse@ritchennai.edu.in
^* sanjay.g.2019.cse@ritchennai.edu.in
^* deepika.n@ritchennai.edu.in

Abstract

The obstacle avoidance system uses a YOLO model to detect obstacles in real-time and provide spatial information about their location and size. This information is then passed to the GTTS system, which generates audio alerts to notify the user of the presence of an obstacle and its location. The audio alerts are generated in a natural-sounding voice to provide the user with clear and concise information. To evaluate the effectiveness of our proposed system, we conducted experiments with visually impaired individuals in real-world scenarios. The results show that our system can significantly improve obstacle detection and avoidance performance compared to traditional methods. The participants reported high levels of satisfaction with the system’s performance and ease of use.