Multilabel Acoustic Event Classification Using Real-World Urban Data and Physical Redundancy of Sensors
Vidaña-Vila,Ester Navarro,Joan Stowell,Dan Alsina-Pagès,Rosa Ma
Vidaña-Vila,Ester
Navarro,Joan
Stowell,Dan
Alsina-Pagès,Rosa Ma
Abstract
Many people living in urban environments nowadays are overexposed to noise, which results in adverse effects on their health. Thus, urban sound monitoring has emerged as a powerful tool that might enable public administrations to automatically identify and quantify noise pollution. Therefore, identifying multiple and simultaneous acoustic sources in these environments in a reliable and cost-effective way has emerged as a hot research topic. The purpose of this paper is to propose a two-stage classifier able to identify, in real time, a set of up to 21 urban acoustic events that may occur simultaneously (i.e., multilabel), taking advantage of physical redundancy in acoustic sensors from a wireless acoustic sensors network. The first stage of the proposed system consists of a multilabel deep neural network that makes a classification for each 4-s window. The second stage intelligently aggregates the classification results from the first stage of four neighboring nodes to determine the final classification result. Conducted experiments with real-world data and up to three different computing devices show that the system is able to provide classification results in less than 1 s and that it has good performance when classifying the most common events from the dataset. The results of this research may help civic organisations to obtain actionable noise monitoring information from automatic systems.
Description
Funding Information: Funding: We would like to thank Secretaria d’Universitats i Recerca of the Department d’Empresa i Coneixement of the Generalitat de Catalunya for partially funding this work under grants 2017-SGR-966 and 2017-SGR-977. Additionally, we would like to thank La Salle Campus BCN-URL for partially funding the joint research with Tilburg University in the framework of Ms. Vidaña-Vila’s PhD thesis. Publisher Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
Date
2021-11-10
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Keywords
Acoustic event classification, Deep neural networks, Distributed computing, Multilabel classification, Physical redundancy, Urban sound monitoring, SDG 3 - Good Health and Well-being, SDG 11 - Sustainable Cities and Communities
Citation
Vidaña-Vila, E, Navarro, J, Stowell, D & Alsina-Pagès, R M 2021, 'Multilabel Acoustic Event Classification Using Real-World Urban Data and Physical Redundancy of Sensors', Sensors, vol. 21, no. 22, 7470. https://doi.org/10.3390/s21227470