MVD:A Novel Methodology and Dataset for Acoustic Vehicle Type Classification

TL;DR

This paper introduces MVD and MVDA, two datasets for acoustic vehicle classification, and proposes a neural network approach that achieves high accuracy, demonstrating the effectiveness of acoustic monitoring for traffic management.

Contribution

The paper presents new open datasets and a novel neural network method for vehicle-type classification using acoustic signals, improving accuracy over previous approaches.

Findings

Achieved 91.98% accuracy on MVD dataset.

Achieved 96.66% accuracy on MVDA dataset.

Deployed the model in an Android app for practical testing.

Abstract

Rising urban populations have led to a surge in vehicle use and made traffic monitoring and management indispensable. Acoustic traffic monitoring (ATM) offers a cost-effective and efficient alternative to more computationally expensive methods of monitoring traffic such as those involving computer vision technologies. In this paper, we present MVD and MVDA: two open datasets for the development of acoustic traffic monitoring and vehicle-type classification algorithms, which contain audio recordings of moving vehicles. The dataset contain four classes- Trucks, Cars, Motorbikes, and a No-vehicle class. Additionally, we propose a novel and efficient way to accurately classify these acoustic signals using cepstrum and spectrum based local and global audio features, and a multi-input neural network. Experimental results show that our methodology improves upon the established baselines of previous works and achieves an accuracy of 91.98% and 96.66% on MVD and MVDA Datasets, respectively. Finally, the proposed model was deployed through an Android application to make it accessible for testing and demonstrate its efficacy.