Anomalous sound detection (ASD) is an important technology in the fourth industrial revolution, which can monitor the abnormal state of machine by identifying whether the sound of the machine is normal or not. However, in practical applications where there are few anomalous sound samples from machines, achieving effective ASD is still a challenge. In this paper, an unsupervised ASD algorithm based on adversarial autoencoder with attention mechanism is proposed. Different from the traditional reconstruction-based ASD model, in order to make the features learned by the model more representative, complex sound timing signals are converted into Gammatone spectrogram with richer features through filtering. Then the spectrogram is used as the input of the convolutional autoencoder. At the same time, the attention mechanism is introduced in the encoder to enhance adaptive learning of the normal patterns. Then the discriminator is used in the generative adversarial network to perform adversarial learning with the improved convolutional autoencoder to improve the reconstruction ability of the model for normal samples. Experimental results demonstrate that the proposed algorithm significantly outperforms commonly used industry methods for anomaly detection and exhibits advantages over other deep learning approaches in terms of system complexity.
PurarjomandlangrudiAGhapanchiAHEsmalifalakM. A data mining approach for fault diagnosis: an application of anomaly detection algorithm[J]. Measurement ( Mahwah N J)2014; 55: 343–352.
2.
Blázquez-GarcíaACondeAMoriU, et al.A review on outlier/anomaly detection in time series data[J]. ACM Comput Surveys (CSUR)2021; 54: 1–33.
3.
BergmannPFauserMSattleggerD, et al.MVTec AD–A comprehensive real-world dataset for unsupervised anomaly detection[C]. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019: 9592–9600.
4.
MeiSYangHYinZ. An unsupervised-learning-based approach for automated defect inspection on textured surfaces[J]. IEEE Trans Instrum Meas2018; 67: 1266–1277.
5.
ShuXBaoTZhouY, et al.Unsupervised dam anomaly detection with spatial–temporal variational autoencoder[J]. Structural Health Monit2023; 22: 39–55.
6.
ThillMKonenWWangH, et al.Temporal convolutional autoencoder for unsupervised anomaly detection in time series[J]. Appl Soft Comput2021; 112: 107751.
7.
ZhangLRenLShiY. A humanoid method for extracting abnormal engine sounds from engine acoustics based on adaptive volterra filter[J]. J Bionic Eng2012; 9: 262–270.
8.
YangYWGuanYGChenSG, et al.Mechanical fault diagnosis method of high voltage circuit breaker based on sound signal[J]. Proc CSEE2018; 38: 6730–6737.
9.
BansalSSahooSTiwariR, et al.Multiclass fault diagnosis in gears using support vector machine algorithms based on frequency domain data[J]. Measurement ( Mahwah N J)2013; 46: 3469–3481.
10.
ZhaoXWangDL. Analyzing noise robustness of MFCC and GFCC features in speaker identification[C]. In: 2013 IEEE international conference on acoustics, speech and signal processing. 2013: 7204–7208.
11.
IslamRAbdel-RaheemETariqueM. A novel pathological voice identification technique through simulated cochlear implant processing systems[J]. Appl Sci2022; 12: 2398.
12.
KadriFHarrouFChaabaneS, et al.Seasonal ARMA-based SPC charts for anomaly detection: application to emergency department systems[J]. Neurocomputing2016; 173: 2102–2114.
13.
LiaoNLiX. Traffic anomaly detection model using k-means and active learning method[J]. Int J Fuzzy Syst2022; 24: 2264–2282.
14.
LaxhammarRFalkmanGSviestinsE. Anomaly detection in sea traffic-a comparison of the Gaussian mixture model and the kernel density estimator[C]. In: 2009 12th international conference on information fusion. 2009: 756–763.
15.
PaffenrothRDu ToitPNongR, et al.Space-time signal processing for distributed pattern detection in sensor networks[J]. IEEE J Sel Top Signal Process2013; 7: 38–49.
16.
HoffmannH. Kernel PCA for novelty detection[J]. Pattern Recognit2007; 40: 863–874.
17.
PangJPuXLiC. A hybrid algorithm incorporating vector quantization and one-class support vector machine for industrial anomaly detection[J]. IEEE Trans Ind Inf2022; 18: 8786–8796.
18.
LiuBXiaoYCaoL, et al.SVDD-based outlier detection on uncertain data[J]. Knowl Inf Syst2013; 34: 597–618.
19.
BöttgerTUlrichM. Real-time texture error detection on textured surfaces with compressed sensing[J]. Pattern Recognit Image Anal2016; 26: 88–94.
20.
CarreraDManganiniFBoracchiG, et al.Defect detection in SEM images of nanofibrous materials[J]. IEEE Trans Ind Inf2016; 13: 551–561.
21.
BergmannPFauserMSattleggerD, et al.Uninformed students: student-teacher anomaly detection with discriminative latent embeddings[C]. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 4183–4192.
22.
NapoletanoPPiccoliFSchettiniR. Anomaly detection in nanofibrous materials by CNN-based self-similarity[J]. Sensors2018; 18: 209.
23.
BergmannPLöweSFauserM, et al.Improving unsupervised defect segmentation by applying structural similarity to autoencoders[J]. arXiv preprint arXiv:1807.02011, 2018.
24.
ZavrtanikVKristanMSkočajD. Reconstruction by inpainting for visual anomaly detection[J]. Pattern Recognit2021; 112: 107706.
25.
ParkDHoshiYKempCC. A multimodal anomaly detector for robot-assisted feeding using an LSTM-based variational autoencoder[J]. IEEE Robotics and Automation Letters2018; 3: 1544–1551.
26.
ChandraAKalaR. Regularised encoder-decoder architecture for anomaly detection in ECG time signals[C]. In: 2019 IEEE Conference on Information and Communication Technology. IEEE, 2019: 1–6.
27.
SchleglTSeeböckPWaldsteinSM, et al.Unsupervised anomaly detection with generative adversarial networks to guide marker discovery[C]. In: International conference on information processing in medical imaging. Cham: Springer International Publishing, 2017: 146–157.
28.
SchleglTSeeböckPWaldsteinSM, et al.f-AnoGAN: fast unsupervised anomaly detection with generative adversarial networks[J]. Med Image Anal2019; 54: 30–44.
29.
ZenatiHFooCSLecouatB, et al.Efficient gan-based anomaly detection[J]. arXiv preprint arXiv:1802.06222, 2018.
30.
AkcaySAtapour-AbarghoueiABreckonTP. Ganomaly: Semi-supervised anomaly detection via adversarial training[C]//Computer Vision–ACCV 2018: 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part III 14. Springer International Publishing, 2019: 622-637.
31.
JohannesmaPIM. The pre-response stimulus ensemble of neurons in the cochlear nucleus[J]. 1972, pp. 58–69.
LinJ. Feature extraction of machine sound using wavelet and its application in fault diagnosis[J]. NDT E Int2001; 34: 25–30.
34.
HuangHBLiRXHuangXR, et al.Sound quality evaluation of vehicle suspension shock absorber rattling noise based on the Wigner–Ville distribution[J]. Appl Acoust2015; 100: 18–25.
35.
ZouFLiXLiY, et al.GOL-SFSTS based few-shot learning mechanical anomaly detection using multi-channel audio signal[J]. Knowl Based Syst2024; 284: 111204.
36.
WooSParkJLeeJY, et al.CBAM: convolutional block attention module[C]. In: Proceedings of the European conference on computer vision (ECCV). 2018: 3–19.
37.
GoodfellowIPouget-AbadieJMirzaM, et al.Generative adversarial nets[J]. Adv Neural Inf Process Syst2014: 27: 1–9.
38.
SalimansTGoodfellowIZarembaW, et al.Improved techniques for training GANs[J]. Adv Neural Inf Process Syst2016: 29: 1–9.
39.
TanabeRPurohitHDohiK, et al.MIMII DUE: sound dataset for malfunctioning industrial machine investigation and inspection with domain shifts due to changes in operational and environmental conditions[C]//2021 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). IEEE, 2021: 21–25.
40.
HaradaNNiizumiDTakeuchiD, et al.ToyADMOS2: Another dataset of miniature-machine operating sounds for anomalous sound detection under domain shift conditions[J]. arXiv preprint arXiv:2106.02369, 2021.
41.
YangJShiYQiZ. Learning deep feature correspondence for unsupervised anomaly detection and segmentation[J]. Pattern Recognit2022; 132: 108874.
42.
GiriRTennetiSVHelwaniK, et al.Unsupervised anomalous sound detection using self-supervised classification and group masked autoencoder for density estimation[J]. Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2020 Challenge), Tech. Rep, 2020, 23.
43.
DanilukPGoździewskiMKapkaS, et al.Ensemble of auto-encoder based and Wavenet like systems for unsupervised anomaly detection[J]. Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2020 Challenge), Tech. Rep, 2020.
44.
PrimusP. Reframing unsupervised machine condition monitoring as a supervised classification task with outlier-exposed classifiers[J]. Tech. Rep., DCASE2020 Challenge, 2020.
45.
InoueTVinayavekhinPMorikuniS, et al.Detection of Anomalous Sounds for Machine Condition Monitoring using Classification Confidence[C]//DCASE. 2020: 66–70.
