Real Time Assistance for Visually Impaired using Text and Object Recognition

For blind people, visual impairment substantially impedes daily tasks. In order to improve users spatial awareness and information access, this project suggests creating smart glasses with text and object detecting features in addition to voice pronunciation. For real-time object detection of frequently observed things, such as people, phones, tables, and animals, the system makes use of You Only Look Once (YOLO). Furthermore, Google Text-to-Speech enables text capture and conversion into voice through the use of Optical Character Recognition (OCR) technology. Two user-selectable modes are available for the smart glasses are OCR and ODM. In OCR Mode, the camera to take a still image, OCR the text, and then speak the text that has been retrieved. In Object Detection Mode, continuously finds and vocally announces items in the user’s field of vision. The goal of this initiative is to increase the safety and independence.