Abstract— This project presents the design and development of an advanced smart assistance system for visually impaired individuals, leveraging mobile-based artificial intelligence to provide real time obstacle detection and navigation support. Unlike traditional systems that rely heavily on ultrasonic sensors and physical hardware, this solution utilizes the camera of a smartphone combined with AI- powered object detection models to identify and classify nearby objects. The detected objects are then translated into meaningful audio feedback using text-to- speech technology, informing the user about the object type and its relative direction (e.g., "Person on your left" or "Car in front"). The goal is to offer an affordable, scalable, and user- friendly alternative to hardware-heavy solutions while enhancing mobility and independence. By eliminating the need for complex sensor arrays and instead using a mobile phone, the system ensures accessibility and portability. The proposed application not only processes live video streams efficiently but also delivers timely alerts, helping the user avoid obstacles and navigate safely in dynamic environments. Additionally, the app supports multilingual audio feedback, which caters to a diverse user base. This work aims to bridge the gap between low cost solutions and high-functionality requirements in assistive technologies, offering a powerful tool for improving the quality of life for visually impaired users.
PADMINI et al. (Mon,) studied this question.