Object Detection Tracking and Alert System for Visually Impaired Persons

M. Joshna, J. Tharun Kumar, G. Yuvaraju, K. Vishnu Vardhan Reddy, V. Sai Prathap Reddy

2025

Abstract

This project helps the blind in their navigation using advanced computer vision and speech synthesis technology. Based on the COCO dataset and a combination of SSD MobileNet V3 model with OpenCV's DNN module, the system detects the mentioned objects in real-time from webcam feeds. The identified objects are marked with bounding boxes with tags and pyttsx3 is used to convert these text tags to speech alerts, allowing real-time auditory indicators of what the user’s surroundings are. The SSD MobileNet V3 model was selected for use in this project, as it represented the optimal trade-off between accuracy and computation – perfect for implementation in real-time applications on edge devices. They use the techniques of object detection and speech synthesis. It applies the detection algorithm on every frame of the video, draws a box around the detected object on the frame and activates a speech alert with the detected object name by utilizing pyttsx3. It also offers users up-to-date, succinct information on objects in their environment and improving their situational awareness and security.

Paper Citation

in Harvard Style

Joshna M., Kumar J., Yuvaraju G., Reddy K. and Reddy V. (2025). Object Detection Tracking and Alert System for Visually Impaired Persons. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 45-49. DOI: 10.5220/0013876700004919

in Bibtex Style

@conference{icrdicct`2525,
author={M. Joshna and J. Kumar and G. Yuvaraju and K. Reddy and V. Reddy},
title={Object Detection Tracking and Alert System for Visually Impaired Persons},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={45-49},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013876700004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Object Detection Tracking and Alert System for Visually Impaired Persons
SN - 978-989-758-777-1
AU - Joshna M.
AU - Kumar J.
AU - Yuvaraju G.
AU - Reddy K.
AU - Reddy V.
PY - 2025
SP - 45
EP - 49
DO - 10.5220/0013876700004919
PB - SciTePress