BSR (Buzz,  Squeak,  Rattle) noise classification based on convolutional neural network with short-time Fourier transform noise-map

Seok-Jun Bu; Se-Min Moon; Sung-Bae Cho

doi:10.7776/ASK.2018.37.4.256

All Issue

2018 Vol.37, Issue 4 Preview Page

Research Article

BSR (Buzz, Squeak, Rattle) noise classification based on convolutional neural network with short-time Fourier transform noise-map Short-time Fourier transform 소음맵을 이용한 컨볼루션 기반 BSR (Buzz, Squeak, Rattle) 소음 분류

31 July 2018. pp. 256-261

PDF XML

Abstract

There are three types of noise generated inside the vehicle: BSR (Buzz, Squeak, Rattle). In this paper, we propose a classifier that automatically classifies automotive BSR noise by using features extracted from deep convolutional neural networks. In the preprocessing process, the features of above three noises are represented as noise-map using STFT (Short-time Fourier Transform) algorithm. In order to cope with the problem that the position of the actual noise is unknown in the part of the generated noise map, the noise map is divided using the sliding window method. In this paper, internal parameter of the deep convolutional neural networks is visualized using the t-SNE (t-Stochastic Neighbor Embedding) algorithm, and the misclassified data is analyzed in a qualitative way. In order to analyze the classified data, the similarity of the noise type was quantified by SSIM (Structural Similarity Index) value, and it was found that the retractor tremble sound is most similar to the normal travel sound. The classifier of the proposed method compared with other classifiers of machine learning method recorded the highest classification accuracy (99.15 %).

Keywords

Automotive noise

BSR (Buzz, Squeak, Rattle) classification

Deep learning

Convolutional neural network

차량 내부에는 BSR(Buzz, Squeak, Rattle) 세 가지 유형의 소음이 발생한다. 본 논문에서는 심층 컨볼루션 신경망으로 추출한 소음 특징에 기반하여 자동으로 차량 내부의 BSR 소음을 분류하는 분류기를 제안한다. 차량 내부의 소음은 전처리 단계에서 STFT(Short-time Fourier Transform) 알고리즘을 사용하여 소음 맵으로 표현된다. 생성된 소음 맵 내부에서 실제 소음의 위치를 정확하게 파악하기 어려운 문제에 대처하기 위해서 슬라이딩 윈도우 방법으로 분할하였다. 본 논문에서는 t-SNE(t-Stochastic Neighbor Embedding) 알고리즘을 사용하여 심층 컨볼루션 신경망 내부 파라미터를 시각화하고 정성적인 방식으로 오분류데이터를 분석하였다. 분류된 데이터의 정량적인 분석을 위해 소음의 종류별 유사도를 SSIM(Structural Similarity Index) 수치에 기반하여 정량화하여 리트랙터의 떨림음이 정상주행음과 가장 유사하다는 것을 밝혔다. 제안하는 방법의 분류기는 기타 기계학습 알고리즘 대비 최고 분류 정확도를 달성하였다(99.15%).

키워드

차량 소음

BSR (Buzz, Squeak, Rattle) 분류

딥 러닝

컨볼루션 신경망

References

M. G. Mog, B. H. Min, S. W. Choi, and H. J. Lee, “Development of the reproduction test method of automobile buzz, squeak, rattle noise and the tracking system” (in Korean), Proc. KSAE 2010 Conference, 1475-1481 (2010).

H. Wu, M. Siegel, and P. Khosla, “Vehicle sound signature recognition by frequency vector principal component analysis,” IMTC, IEEE, 1, 429-434 (1998).

K. C. Kim, S. W. Lee, J. M. Choi, and Y. J. Kim, “Study on the simulation process for the buzz, squeak and rattle incorporating vehicle to vehicle variance” (in Korean), Proc. KSNVE 2014 Conference, 480-481 (2014).

S. U. Choi, M. G. Mog, S. J. Kang, K. D. Yee, and M. W. Suh, “A study on the BSR performance prediction technology” (in Korean), Proc. KSAE 2013 Conference, 2040-2044 (2013).

C. Woo and H. Park, “Measurement of BSR noise of cockpit module materials for automobiles,” Proc. AIP Conference, 1727 (2016).

10.1063/1.4945979

G. Cerrato, “Automotive sound quality-accessories, BSR and brakes,” Sound and Vibration, 43, 10, (2009).

Y. S. Wang, C. M. Lee, D. G. Kim, and Y. Xu, “Sound-quality prediction for nonstationary vehicle interior noise based on wavelet pre-processing neural network model,” J. Sound and Vibration, 299, 933-947 (2007).

10.1016/j.jsv.2006.07.034

S. Lee, Y. Kwak, B. Kim, J. Lee, and J. Park, “Parameter quantification for evaluation of vehicle’s impulsive bsr noise,” Proc. Inter-noise Conference, 3361-3363 (2014).

A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Advances in Neural Information Processing Systems, 1097-1105 (2012).

M. D. Zeiler and R. Fergus, “Visualizing and understanding convolutional networks,” Proc. European Conference on Computer Vision, 818-833 (2014).

10.1007/978-3-319-10590-1_53

N. Srivastava, G. E. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: simple way to prevent neural networks from overfitting,” J. Machine Learning Research, 15, 1929-1958 (2014).

T. G. Dietterich, “Machine learning for sequential data: a review,” Joint IAPR International Workshops on Statistical Techniques in Pattern Recognition and Structural and Syntactic Pattern Recognition, 15-30 (2002).

L. V. D. Maaten and G. Hinton, “Visualizing data using t-SNE,” J. Machine Learning Research, 9, 2579-2605 (2008).

T. N. Sainath, A. R. Mohamed, B. Kingbury, and B. Ramabhadran, “Deep convolutional neural networks for LVCSR,” Proc. ICASSP, 8614-8618 (2013).

10.1109/ICASSP.2013.6639347

S. J. Bu and S. B. Cho, “A hybrid system of deep learning and learning classifier system for database intrusion detection," Proc. HAIS, 615-625 (2017).

10.1007/978-3-319-59650-1_52

Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. on Image Processing, 13, 600-612 (2004).

10.1109/TIP.2003.819861

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 37
No :4
Pages :256-261
Received Date : 2018-06-01
Accepted Date : 2018-07-27
DOI :https://doi.org/10.7776/ASK.2018.37.4.256

The Journal of the Acoustical Society of KoreaISSN:1225-4428(Print) 2287-3775(Online)한국음향학회

All Issue