Smart Wearable Glasses for Real-Time Object and Text Recognition to Assist Visually Impaired Users in Bolivia

Abstract

In Bolivia, more than 120,000 people live with complete blindness and more than 300,000 experience low vision, mostly in underserved regions and moreover with limited access to ophthalmological care. This study presents a developed low-cost prototype of wearable smart glasses with a 3D-printed enclosure, integrating real-time object detection and OCR capabilities. The system is built using off-the-shelf components, including a Raspberry Pi 4 and a digital camera, then for the programming utilizes open-source libraries such as OpenCV, YOLOv4-Tiny, and Tesseract OCR. The device provides auditory feedback in Spanish, English, and basic Aymara, enhancing linguistic accessibility. A mixed-methods evaluation was conducted, combining technical performance metrics and user experience assessments with participants from the Bolivian Institute for the Blind (IBC). Results show high user satisfaction (average rating: 9/10) and a positiveve impact on orientation and autonomy. The solution is inclusive, replicable, and holds potential to improve the quality of life for visually impaired individuals in low-resource settings.