Analysis of Real Time Road Surface and Acoustic Data Processing for Minimizing the Accident Rate Using Feed Forward Neural Network

Eswaramoorthi R., Velliangiri A., Devaviknesh S., Haribaskar S., Naveen S., Prasanth D.

2025

Abstract

Simplified road condition monitoring is essential to maintaining road safety and maximizing transportation efficiency in smart cities. Effective road surface detection has been significantly improved by the use of artificial intelligence (AI). Issues with asphalt pavement are the main concern of both developed and emerging countries for the efficient functioning of everyday commutes. The identification of potholes, which are dangerous to cars and people and can result in an accident, has been the subject of several research. In order to identify potholes on edge devices, this study aims to explore the possibilities of deep learning models and use three outstanding deep learning models. This article proposes a low-cost technology for detecting the surface qualities of road pavement in real-time. The time-frequency domain processing of the inertial signals given by on-car sensors is done in order to get information about the condition of the road surface. The effectiveness of the suggested approach in determining the kind and existence of distress is demonstrated by the high categorization rates. Following data collection from the road surfaces, machine learning methods like Multi-Layer Perceptron (MLP) are used for analysis. The outcomes show how well the suggested approach can distinguish between various road conditions. These findings showed that the MLP had a higher accuracy of 98.98% when evaluating road conditions. In order to give safe transportation services in smart cities, the study offers important insights into the creation of a more effective and dependable road condition monitoring system.

Paper Citation

in Harvard Style

R. E., A. V., S. D., S. H., S. N. and D. P. (2025). Analysis of Real Time Road Surface and Acoustic Data Processing for Minimizing the Accident Rate Using Feed Forward Neural Network. 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 208-213. DOI: 10.5220/0013880200004919

in Bibtex Style

@conference{icrdicct`2525,
author={Eswaramoorthi R. and Velliangiri A. and Devaviknesh S. and Haribaskar S. and Naveen S. and Prasanth D.},
title={Analysis of Real Time Road Surface and Acoustic Data Processing for Minimizing the Accident Rate Using Feed Forward Neural Network},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={208-213},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013880200004919},
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 - Analysis of Real Time Road Surface and Acoustic Data Processing for Minimizing the Accident Rate Using Feed Forward Neural Network
SN - 978-989-758-777-1
AU - R. E.
AU - A. V.
AU - S. D.
AU - S. H.
AU - S. N.
AU - D. P.
PY - 2025
SP - 208
EP - 213
DO - 10.5220/0013880200004919
PB - SciTePress