Detecting Road Surface Wetness Using Microphones and Convolutional Neural Networks

Pepe G, Gabrielli L, Ambrosini L, Squartini S, Cattani L. Detecting Road Surface Wetness Using Microphones and Convolutional Neural Networks. In: AES Convention 146. Audio Engineering Society; 2019. Paper 10193. Available from: https://aes.org/publications/elibrary-page/?id=20326

@inproceedings{Pepe2019_20326,
  author = {Pepe, Giovanni and Gabrielli, Leonardo and Ambrosini, Livio and Squartini, Stefano and Cattani, Luca},
  title = {{Detecting Road Surface Wetness Using Microphones and Convolutional Neural Networks}},
  booktitle = {AES Convention 146},
  note = {Paper 10193},
  year = {2019},
  month = mar,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=20326}
}

TY  - CPAPER
TI  - Detecting Road Surface Wetness Using Microphones and Convolutional Neural Networks
AU  - Pepe, Giovanni
AU  - Gabrielli, Leonardo
AU  - Ambrosini, Livio
AU  - Squartini, Stefano
AU  - Cattani, Luca
T2  - AES Convention 146
M1  - Paper 10193
PY  - 2019
DA  - 2019/03/06
UR  - https://aes.org/publications/elibrary-page/?id=20326
PB  - Audio Engineering Society
LA  - en
AB  - The automatic detection of road conditions in next-generation vehicles is an important task that is getting increasing interest from the research community. Its main applications concern driver safety, autonomous vehicles, and in-car audio equalization. These applications rely on sensors that must be deployed following a trade-off between installation and maintenance costs and effectiveness. In this paper we tackle road surface wetness classification using microphones and comparing convolutional neural networks (CNN) with bi-directional long-short term memory networks (BLSTM) following previous motivating works. We introduce a new dataset to assess the role of different tire types and discuss the deployment of the microphones. We find a solution that is immune to water and sufficiently robust to in-cabin interference and tire type changes. Classification results with the recorded dataset reach a 95% F-score and a 97% F-score using the CNN and BLSTM methods, respectively.
ER  -