Smart Glass for Visually Impaired Individuals Using ML and IOT

This paper adopts the YOLO (You Only Look Once) algorithm for integrating machine learning-based object detection into the second stage of the Smart Glass System designed for visually impaired people. Assistive technology requires object detection because it enables users to enhance their spatial awareness and avoid obstacles and execute selfnavigation. The research introduces an indoor detection system which recognizes objects in real time while generating audio feedback using speakers or Bluetooth modulations. This system depends on a microcontroller to handle processing work and features both a camera module for video recording alongside a YOLO model with pre-trained accuracy capability for object classification. Real-time operational assistance occurs automatically without requiring external connectivity because the system speaks detected object names to users at the moment of detection. The proposed system functions by detecting multiple objects simultaneously which enhances the situational understanding of visually impaired users in dynamic indoor environments. The method demonstrates its detection precision together with response time and adaptability features through testing it in various real-world indoor locations. The experimental findings indicate that YOLO serves as an appropriate tool for real-time assistive systems because it delivers rapid computations and excellent object detection abilities. This research enables the development of assistive technologies built by artificial intelligence which links visual impairment accessibility to computer vision solutions while promoting safety and independence.