Research on Vehicle Classification and Recognition Method Based on Vehicle Acoustic Signal CNN Analysis

DOI

10.2991/seeie-19.2019.61How to use a DOI?

Keywords

intelligent transportation; vehicle classification recognition; vehicle acoustic signal; feature extraction; deep learning; convolution neural network

Abstract

The present "shallow classification model" have shortcomings on modeling and representation ability, feature extraction, classification performance and so on. This study aims to improve the typical LeNet-5 convolution neural network and obtain three kinds of CNN structures to realize the classification of large and small vehicles. Firstly, we extracted the MFCC feature of vehicle acoustic signals; then took the feature signals as training samples; lastly adjusted the study rate, convolution kernel size and quantity in accordance with experiment and obtained the results. The experimental results indicate that the improved CNN model is better than the traditional machine learning method; and the classification performance of the improved CNN model is improved with the increase of data volume, and the accuracy of the test samples is 96.8%.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhangli Lan
AU  - Yuxin Zhang
AU  - Juan Cao
AU  - Ranran Tang
AU  - Liyun Tan
AU  - Fang Liu
PY  - 2019/05
DA  - 2019/05
TI  - Research on Vehicle Classification and Recognition Method Based on Vehicle Acoustic Signal CNN Analysis
BT  - Proceedings of the 2019 2nd International Conference on Sustainable Energy, Environment and Information Engineering (SEEIE 2019)
PB  - Atlantis Press
SP  - 263
EP  - 266
SN  - 2352-5401
UR  - https://doi.org/10.2991/seeie-19.2019.61
DO  - 10.2991/seeie-19.2019.61
ID  - Lan2019/05
ER  -