About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I

Research Article

Real-Time Object Detection and Audio Feedback Wearable System for Visually Impaired Using Raspberry Pi

Download17 downloads
Cite
BibTeX Plain Text
  • @INPROCEEDINGS{10.4108/eai.28-4-2025.2357925,
        author={Lakshana.  M and Keren.  J and Priyadharshini.  R and Keerthana.  M and Karthikeyan.  A},
        title={Real-Time Object Detection and Audio Feedback Wearable System for Visually Impaired Using Raspberry Pi},
        proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I},
        publisher={EAI},
        proceedings_a={ICITSM PART I},
        year={2025},
        month={10},
        keywords={raspberry pi object detection yolov3 visual impairment assistive technology},
        doi={10.4108/eai.28-4-2025.2357925}
    }
    
  • Lakshana. M
    Keren. J
    Priyadharshini. R
    Keerthana. M
    Karthikeyan. A
    Year: 2025
    Real-Time Object Detection and Audio Feedback Wearable System for Visually Impaired Using Raspberry Pi
    ICITSM PART I
    EAI
    DOI: 10.4108/eai.28-4-2025.2357925
Lakshana. M1,*, Keren. J1, Priyadharshini. R1, Keerthana. M1, Karthikeyan. A1
  • 1: Mahendra Engineering College
*Contact email: lakshanaiitm@gmail.com

Abstract

People who are visually impaired encounter regular obstacles when trying to detect and move around their surroundings. The research presents "Third Eye" which represents a cost-effective wearable system for blind users that uses Raspberry Pi 4 as its operating platform. Through its camera the system obtains live visual inputs which leads to object detection processing with YOLOv3, Haar Cascade and SSD before issuing audio feedback through Google Speech API. This system based on OpenCV technology within Python works to convert surrounding visuals into descriptive audio output which improves user independence and awareness. The evaluation demonstrates that YOLOv3 performs object detection with accuracy rates spanning 80% to 99% which results in a mean Average Precision (mAP) of 31.05% while processing frames that operate at speeds from 10.12 to 16.29 FPS for real-time applications. The system stands as an affordable substitute to advanced assistive equipment because it enables visually impaired users to improve their lifestyle.

Keywords
raspberry pi, object detection, yolov3, visual impairment, assistive technology
Published
2025-10-13
Publisher
EAI
http://dx.doi.org/10.4108/eai.28-4-2025.2357925
Copyright © 2025–2025 EAI
EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL