Object Detection for Blind Person

Abstract

Independence is a fundamental right for every individual, including those living with visual impairments. With rapid advancements in technology over the past decades, significant efforts have been made to develop tools that assist disabled individuals in achieving autonomy. This study introduces a smart assistive system specifically designed to enhance the mobility of blind individuals by providing real-time information about their surroundings. The proposed solution employs the You Only Look Once (YOLO) object detection algorithm for high-speed and accurate identification of various objects within video streams. The system is implemented using OpenCV and Python and operates on the Raspberry Pi 3 platform, a low-cost and portable computing device. Results from the implementation show that the system is capable of recognizing multiple types of objects with high accuracy and delivering audio-based feedback to the user. This enables the visually impaired to navigate both indoor and outdoor environments with greater confidence and safety.