Title :
Environmental Noise Source Classification Using Neural Networks
Author :
Barkana, Buket D. ; Saricicek, Inci
Author_Institution :
Dept. of Electr. Eng., Univ. of Bridgeport, Bridgeport, CT, USA
Abstract :
Neural networks have been applied to many interesting problems in different areas including noise identification/recognition. With this study, we studied noise classification using artificial neural networks (ANN). Three commonly encountered non-stationary noise sources are chosen to recognize. These are highway, subway and airport. Time-domain based feature parameters are used. While one-phase ANN classifier achieving 54% accuracy, two-phase ANN classifier achieved 83-89% accuracy rates.
Keywords :
acoustic signal processing; neural nets; noise (working environment); pattern classification; ANN classifier; airport noise; artificial neural networks; environmental noise; highway noise; noise source classification; nonstationary noise sources; subway noise; time-domain based feature parameters; Acoustic noise; Artificial neural networks; Background noise; Classification tree analysis; Electronic mail; Hidden Markov models; Neural networks; Pattern recognition; Speech; Working environment noise; ACF-based feature parameter; Neural Networks (ANN); environmental noise classification;
Conference_Titel :
Information Technology: New Generations (ITNG), 2010 Seventh International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-6270-4
DOI :
10.1109/ITNG.2010.118
