Salta al menu principale di navigazione Salta al contenuto principale Salta al piè di pagina del sito

Articoli Scientifici

V. 47 N. 2 (2023)

Metodi di deep learning acustico per il riconoscimento dei dissesti della pavimentazione stradale

DOI
https://doi.org/10.3280/ria2-2023oa15509
Inviata
28 febbraio 2023
Pubblicato
14-02-2024

Abstract

Nel seguente lavoro è stata proposta una metodologia basata su tecniche di deep learning per la valutazione delle condizioni della superficie stradale a partire da segnali acustici misurati all’interno della cavità dello pneumatico. Il progetto è stato svolto in collaborazione con Ipool srl., nel contesto del progetto SURFAce, finanziato dalla regione Toscana. Sono state proposte tre architetture di classificazione: una LSTM (Long short-term memory network) basata sull’andamento temporale di un insieme di descrittori spettrali e due CNN (Convolutional neural network), una incentrata sugli spettrogrammi dei segnali, l’altra sui Mel-frequency cepstral coefficients (MFCC). Il dataset di ground truth è stato acquisito tramite un laboratorio mobile e classificato mediante strumenti di analisi appositamente sviluppati. Due delle tre architetture proposte hanno fornito risultati incoraggianti. L’implementazione di tali strumenti su dispositivi mobili potrebbe rendere possibile la classificazione dello stato della pavimentazione in tempo reale con ridotti costi economici e temporali.

Riferimenti bibliografici (comprensivi di DOI)

  1. G. Bitelli, A. Simone, F. Girardi, C. Lantieri, Laser scanning on road pavements: A new approach for characterizing surface texture, Sensors 12 (2012) 9110-9128. https://doi.org/10.3390/s120709110.
  2. M.A. Pallas, M. Bérengier, R. Chatagnon, M. Czuka, M. Conter, M. Muirhead, Towards a model for electric vehicle noise emission in
  3. the European prediction method CNOSSOS-EU, Appl. Acoust. 113 (2016) 89-101. https://doi.org/10.1016/j.apacoust. 2016.06.012.
  4. ISO 11819-2:2017 Acoustics - Measurement of the influence of road surfaces on traffic noise - Part 2: The close-proximity method, International Organization for Standardization, Geneva, Switzerland, 2017.
  5. J. Masino, B. Daubner, M. Frey, F. Gauterin, Development of a tire cavity sound measurement system for the application of field operational tests, in: 10th Annual International Systems Conference, SysCon 2016 - Proceedings, Institute of Electrical and Electronics Engineers Inc., Orlando, FL, 2016: 7490624. https://doi.org/10.1109/SYSCON.2016.7490624.
  6. L.G. Del Pizzo, F. Bianco, A. Moro, G. Schiaffino, G. Licitra, Relationship between tyre cavity noise and road surface characteristics on low-noise pavements, Transport. Res. D-Tr. E. 98 (2021) 102971. https://doi.org/10.1016/j.trd.2021.102971.
  7. ISO 13473-3:2002 Acoustics - Characterization of pavement texture by use of surface profiles, International Organization for Standardization, Geneva, Switzerland, 2002.
  8. P. Klein, J.F. Hamet, Road texture and rolling noise: an envelopment procedure for tire-road contact, 2004, 17p. hal00546120.
  9. A. Del Pizzo, Analysis of Tyre Rolling Noise on Low Noise Pavements, PhD Thesis, University of Pisa, Italy, 2021.
  10. J. Pinay, H. J. Unrau, F. Gauterin, Prediction of close-proximity tire-road noise from tire cavity noise measurements using a
  11. statistical approach, Appl. Acoust. 141 (2018) 293-300. https://doi.org/10.1016/j.apacoust.2018.07.023.
  12. G. Schiaffino, L. G. Del Pizzo, S. Silvestri, F. Bianco, G. Licitra, F.G. Pratico, Machine learning techniques applied to road health
  13. status recognition through tyre cavity noise analysis, J. Phys. Conf. Ser. 2162 (2022) 012011. https://doi.org/10.1088/1742-6596/2162/1/012011.
  14. Bollettino ufficiale della regione Lombardia - 1o supplemento straordinario. Allegato B, D.g.r. 25 gennaio 2006 - n. 8/1790 (in
  15. Italian).
  16. S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comp. 9 (1997) 1735-1780. https://doi.org/ 10.1162/neco.1997.9.8.1735.
  17. Z. Li, F. Liu, W. Yang, S. Peng, J. Zhou, A survey of convolutional neural networks: analysis, applications, and prospects, IEEE T. Neur. Net. Lear. 33 (2021) 6999-7019. https://doi.org/10.1109/TNNLS.2021.3084827.
  18. K. Fukushima, S. Miyake, Neocognitron: A self-organizing neural network model for a mechanism of visual pattern recognition, in: Amari, Si., Arbib, M.A. (eds) Competition and Cooperation in Neural Nets. Lecture Notes in Biomathematics, vol 45. Springer,
  19. Berlin, Heidelberg: pp. 267-285. https://doi.org/10.1007/978-3-642-46466-9_18.
  20. S. Davis, P. Mermelstein, Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences, IEEE T. Acoust. Speech. 28 (1980) 357-366. https://doi.org/10.1109/TASSP.1980.1163420.
  21. G. Peeters, A large set of audio features for sound description (similarity and classification) in the CUIDADO project, CUIDADO Project Report, 2004.
  22. A. Lerch, An introduction to audio content analysis: Applications in signal processing and music informatics, Wiley-IEEE Press, Hoboken, NJ, 2012.
  23. E. Scheirer, M. Slaney, Construction and evaluation of a robust multifeature speech/music discriminator, in: Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 5), Institute of Electrical and Electronics Engineers, 1997: pp. 1331-1334.
  24. D.S. Wilks, Statistical methods in the atmospheric sciences, 3rd ed., Academic Press, Oxford, UK; Amsterdam, The Netherlands; Waltham, MA; San Diego, CA, 2011.
  25. S. Raschka, Python machine learning, 2nd ed., Packt Publishing Ltd, Birmingham, UK, 2015.
  26. Y. Wang, S. Ji, H. Xu, Non-stationary signals processing based on STFT, in: 2007 8th International Conference on Electronic Measurement and Instruments, ICEMI, Xi’an, China, 2007: pp. 3301-3304. https://doi.org/10.1109/ICEMI.2007.4350914.
  27. J. Pomerat, A. Segev, R. Datta, On neural network activation functions and optimizers in relation to polynomial regression, in: Proceedings - 2019 IEEE International Conference on Big Data, Big Data 2019, Institute of Electrical and Electronics Engineers Inc., Los Angeles, CA, 2019: pp. 6183-6185. https://doi.org/10.1109/BigData47090.2019.9005674.
  28. Z. Zhang, Improved Adam optimizer for deep neural networks, in: 2018 IEEE/ACM 26th International Symposium on Quality of Service, IWQoS 2018, Institute of Electrical and Electronics Engineers Inc., Banff, Canada, 2018: pp. 1-2. https://doi.org/10.1109/IWQoS.2018.8624183.

Metriche

Caricamento metriche ...