ASSISTIVE DEVICE FOR THE VISUALLY IMPAIRED USING FACE AND OPTICAL CHARACTER RECOGNITION WITH VOICE ACTIVATION

This paper presents an assistive device designed to aid visually impaired individuals through face recognition,optical character recognition (OCR), and voice-activated text-to-speech functionalities. Implemented using theRaspberry Pi Compute Module 4 (CM4), the system integrates Python OpenCV for facial recognition, Tesseractfor OCR, and a google text-to-speech (gTTS) converter for audio feedback. Mono 8MP camera capture imagesfor face and text processing. The deep learning-based face recognition module identifies individuals, while theOCR module extracts and converts text into speech, enabling real-time interaction. The device providespractical solutions for applications such as assistive technologies, smart surveillance, and document processing.We detail the system's design, methodology, implementation, and performance, highlighting challenges andpotential improvements for future iterations.