Low noise surfaces have been increasingly considered as a viable and cost-effective alternative to acoustical barriers. However, road planners and administrators frequently lack information on the correlation between the type of road surface and the resulting noise emission profile. To address this problem, a method to identify and classify different types of road pavements was developed, whereby near field road noise is analyzed using statistical learning methods. The vehicle rolling sound signal near the tires and close to the road surface was acquired by two microphones in a special arrangement which implements the Close-Proximity method. A set of features, characterizing the properties of the road pavement, was extracted from the corresponding sound profiles. A feature selection method was used to automatically select those that are most relevant in predicting the type of pavement, while reducing the computational cost. A set of different types of road pavement segments were tested and the performance of the classifier was evaluated. Results of pavement classification performed during a road journey are presented on a map, together with geographical data. This procedure leads to a considerable improvement in the quality of road pavement noise data, thereby increasing the accuracy of road traffic noise prediction models. (C)
A new method to classify and identify different types of road pavements by analysing the near field sound profile and texture using statistical learning methods is proposed. A set of characteristics were extracted from the noise profile and from the road surface texture. Sound measurements were carried out following the close-proximity method with the texture descriptors being provided by a high speed profilometer system. As a first approach, it is assumed that the features extracted from the noise and texture characteristics follow normal distributions. However, this assumption is not completely verified for all types of road surfaces. The method presented herein exploits the use of Bayesian analysis complemented by a neural network in order to improve the classification results.
The measurement procedure to evaluate the influence of road surface characteristics on vehicle and traffic noise is designated by Close-Proximity (CPX) method, as described in the ISO 11819-2 draft. This procedure consists on acquiring the vehicle rolling noise signal near the tires and close to the surface by means of at least two microphones, in a special arrangement for the determination of the Close-Proximity Sound Index (CPXI). Roadtraffic noise is estimated by taking into account the absorption characteristics of road surface on the propagation of sound and the speed and type of vehicles. However, the particular characteristics of the different pavement types, which may influence the sound radiation, are not considered. The main goal of this research is to identify and classify different types of road pavements, for different stress conditions, using the CPX method. Such information can be used as a guideline for calibrating noise mapping models in order to achieve more realistic and accurate results. The classification of the different road surfaces consists on a supervised learning technique based on the Support Vector Machine, SVM, algorithms. Results based on error analysis are presented and discussed.
Measures aiming environmental noise abatement usually consider acoustic barriers alongside the road. However, the cost associated with these measures is usually considerably high and its performance in urban areas is reduced. The problem of the visual impact is another issue affecting the communities. Nowadays, road planners have started to consider silent surfaces as an alternative. These types of surfaces are constituted basically by changing the texture and/or porosity of the mixtures. In some conditions, noise level abatements up to 15 dB can be achieved. Therefore, a considerable variety of different surfaces are available. The main goal of this research is to identify and classify different types of road pavements by analyzing the noise profile, using the close proximity method. Feature extraction and selection is one of the first procedures on a classifier algorithm. Moreover, the accuracy of the results is strongly dependent on the right choice of the selected feature vector. Standard classifiers are being tested in order to establish guidelines for future developments of this research. In situations of net road surveillance, searching for inhomogeneities on the surface and the presentation of the results in a geographic map, showing the locations of the surface types and the noise levels, can improve the accuracy of the noise mapping models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.