Smart System for Visually Impaired People
International Journal of Creative and Open Research in Engineering and Management · 22.06.2026 · 02 (06) · 1-6
DOI: 10.55041/ijcope.v2i6.279 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 22.06.2026. Verfasst von Sanika Sawale Sanika Sawale und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Visual impairment presents profound chal-lenges to independent navigation, particularly in dynamic urban environments. Smart system For Visually Impaired People is a real time AI-powered mobile application built with Flutter that helps visually impaired indi viduals perceive their surroundings through an ensemble of on-device intelligence mod ules. The system integrates a YOLOv8 model compiled to TensorFlow Lite format for detecting eighty COCO object classes, a Sobel edge-analysis pipeline for recognis ing doors and staircases, Google ML Kit for on-device optical character recognition, a continuous speech recognition listener for voice command input, and a priority-driven text-to-speech engine for natural-language audio output. Every inference pipeline exe cutes locally on the Android device, eliminating dependence on network connectivity and preserving user privacy. A non-maximum suppression stage duplicates overlapping de tections, and a per label cooldown mechanism prevents repetitive announcements while still allowing the user to invoke an instant-repeat command. Hazardous objects detected on the user’s direct path trigger both an emergency voice alert and a haptic vibration, providing redundant warnings. Distance estimation is performed through a per-class bounding-box area calibration model, giving users approximate proximity information in metres.This app demonstrates that affordable hardware with on-device AI can effec tively assist visually impaired users. Index Terms—Deep Learning, YOLOv8, TensorFlow Lite, Flut-ter, Object Detection, Visually Impaired, Assistive Technology, COCO Dataset, Text-to-Speech, Optical Character Recognition, Sobel Edge Detection, On-Device AI
Zitieren
APA
Sawale, S. S. S., Salunkhe, S. S. S., Karale, P. K. P., Pawar, S. P. S., & Patil, P. P. P. (2026). Smart System for Visually Impaired People. International Journal of Creative and Open Research in Engineering and Management, 02(06), 1-6. https://doi.org/10.55041/ijcope.v2i6.279
Vancouver
Sawale SSS, Salunkhe SSS, Karale PKP, Pawar SPS, Patil PPP. Smart System for Visually Impaired People. International Journal of Creative and Open Research in Engineering and Management. 2026;02(06):1-6. doi:10.55041/ijcope.v2i6.279
BibTeX
@article{Sawale2026,
author = {Sawale, Sanika Sawale Sanika and Salunkhe, Srushti Salunkhe Srushti and Karale, Pranali Karale Pranali and Pawar, Shruti Pawar Shruti and Patil, Pradeep Patil Pradeep},
title = {Smart System for Visually Impaired People},
journal = {International Journal of Creative and Open Research in Engineering and Management},
year = {2026},
volume = {02},
number = {06},
pages = {1--6},
doi = {10.55041/ijcope.v2i6.279}
}
EUVIMED