Lightweight Real-Time Object Detection and Navigation System for Visually Impaired Using Deep Learning
Abstract
Visual impairment continues to affect millions of individuals worldwide, limiting their independence and access to essential services. Existing assistive technologies often depend on expensive hardware, continuous internet connectivity, or fragmented functionalities that fail to address the diverse needs of visually impaired users.
This paper presents a lightweight real-time object detection and navigation system designed to assist blind and visually impaired individuals using deep learning technologies. The proposed Android-based application integrates YOLOv8 object detection, Google Gemini-powered scene understanding, optical character recognition, GPS-based emergency alert services, and a voice-driven user interface.
TensorFlow Lite is employed to enable efficient on-device inference for real-time obstacle detection while minimizing computational requirements. The system provides audio-based feedback, allowing users to navigate independently and safely.
Experimental evaluations demonstrate high object detection accuracy, low response latency, and improved user experience. The proposed solution offers an affordable and scalable assistive technology suitable for deployment in developing countries where accessibility resources remain limited.
Citation
Samuel Simutwe & Sesu Ovid. (2026)
Lightweight Real-Time Object Detection and Navigation System for Visually Impaired Using Deep Learning
International Journal of Current Science Research (IJCSR)
e-ISSN: 2454-5422
12(6): 2026: 1–7
License
© 2026 The Author(s). Published by Dr. BGR Publications .
The authors retain copyright of this article.
This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
References
- World Health Organization (WHO), World Report on Vision, Geneva, Switzerland, 2023.
- J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788, 2016.
- G. Jocher et al., “YOLOv8 Documentation and Technical Report,” Ultralytics, 2024.
- Google LLC, “TensorFlow Lite Documentation,” 2024.
- Google DeepMind, “Gemini API Developer Guide,” 2025.
- Microsoft Corporation, “Seeing AI: Artificial Intelligence for Accessibility,” 2024.
- Google LLC, “Lookout App for Android,” 2024.
- Danquah, Nkrumah, and Asare, “Assistive Technology Adoption in Sub-Saharan Africa,” Disability and Rehabilitation: Assistive Technology, vol. 19, no. 2, pp. 115–126, 2024.
- Mbunge, Sigwejo, and Fashoto, “Digital Health and Assistive Technologies in Africa,” Telematics and Informatics, vol. 58, pp. 101–114, 2021.
- Mwale and Zulu, “Accessibility Challenges among Visually Impaired Individuals in Zambia,” African Disability Rights Yearbook, vol. 11, pp. 45–59, 2023.
