Chapter 10 - Using wavelet transformation for acoustic signal processing in heavy vehicle detection and classification

Abstract

The technological era of the Internet of Things and machine learning has had a major impact on the revolution taking place in the logistics and transport sectors. In intelligent vehicle management systems (IVMS), acoustic signal processing plays a vital role in vehicle identification, vehicle condition monitoring, vehicle safety analysis, etc. One of the interesting applications is vehicle identification using acoustic signal processing. The acoustic frequency of each vehicle is significant different. However, the raw acoustic signals are prone to noise and losses, so it is essential to clean these acoustic signals to obtain useful information. Machine learning, based on mathematical models, and deep learning techniques are efficient ways to perform the acoustic signal analysis. This chapter focuses on the different wavelet transform techniques in feature extraction of acoustic signal processing. Here, the feature extraction is carried out based on maximum acoustic signal energy. Further, the chapter analyzes the performance of wavelet transforms and its influence on machine learning models of heavy vehicle detection and classification.