Loading…
This event has ended. Visit the official site or create your own event on Sched.
Welcome to ICSV26!
Back To Schedule
Monday, July 8 • 15:50 - 16:10
ROBUSTNESS ANALYSIS OF NON-NEGATIVE MATRIX FACTORIZATION TECHNIQUES FOR THE ESTIMATION OF URBAN TRAFFIC SOUND LEVELS

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
The advent of low-cost acoustic sensor networks in cities allows the development of innovative approaches such as sound environment recognition, sound source detection, and more. Such tools are capable of improving the quality of the monitoring of the acoustic quality of cities. In order to improve the context of road traffic noise mapping, the use of Non-negative Matrix Factorization (NMF) techniques, as a source separation method to estimate the urban road traffic sound levels from measurements, has proved to be a successful approach. In order to gain knowledge about the robustness of those techniques, this paper studies their behaviors of the initialized thresholded NMF on a corpus composed of urban sound scenes mixing traffic and specific interfering components with calibrated sound. The results reveal the different performances of this approach to estimate the road traffic sound level, depending on the noise levels of the interfering sources, their proximity to the urban traffic spectrum and the choice of the threshold value.


Monday July 8, 2019 15:50 - 16:10 EDT
Westmount 4
  T04 Environ. & commun. noise, RS01 Commun & envirnmt noise