LOW COST SMART GLASSES FOR VISUALLY IMPAIRED USING RASPBERRY PI

Blindness is the inability to perceive light, leading individuals to rely on guiding tools and human assistance. This study introduces a prototype of smart glasses designed to help blind users by detecting objects and text signs, providing audio feedback. The device uses a Raspberry Pi as the microprocessor and a camera for detection. Object detection is powered by the YOLO algorithm, while Optical Character Recognition(OCR) identifies both handwritten and printed text. MATLAB is employed for OCR tasks, which include image capturing, text extraction, and text-to-speech conversion. The prototype can detect up to 15 objects and operates effectively under varying light conditions. The system can use voice responds to visually impaired people about the objects in front of them by uploading the photos to our backend object detection system through the camera function of smart glasses and then download the text descriptions of the result and then use the text-to-speech function . The average processing time for the prototype is about 1.9 seconds while walking quickly and 1.7 seconds at a slower pace.