Smart Harness Design for Blind and Impaired Visually People

This work relates to an artificial vision embedded system intended to improve the autonomy of blind and visually impaired people by providing audio descriptions of texts and surrounding scenes. The proposed Smart Harness offers a new solution for guiding this category of society by introducing technologies that enable them to read text and detect obstacles in their environment. The system uses a Raspberry Pi with a camera to capture images, which are then processed by advanced technologies such as OpenCV to improve image quality, Tesseract OCR and eSpeak for text recognition, and TensorFlow for object detection. The extracted information and descriptions are converted to audio via eSpeak and transmitted to the user through a connected audio device. This system provides real-time audio descriptions, helping visually impaired people navigate their environment, thereby increasing their independence and quality of life.