Deep Autoencoding GMM-based Unsupervised Anomaly Detection in Acoustic Signals and its Hyper-parameter Optimization
This work addresses the need for automatic machine inspection to prevent costly failures, but it is incremental as it adapts an existing method to a new domain with optimizations.
The paper tackled the problem of poor anomaly detection performance in acoustic signals for factory machinery by proposing DAGMM-HO, a method combining deep autoencoding Gaussian mixture models with hyper-parameter optimization, which achieved up to a 20% improvement in AUC score.
Failures or breakdowns in factory machinery can be costly to companies, so there is an increasing demand for automatic machine inspection. Existing approaches to acoustic signal-based unsupervised anomaly detection, such as those using a deep autoencoder (DA) or Gaussian mixture model (GMM), have poor anomaly-detection performance. In this work, we propose a new method based on a deep autoencoding Gaussian mixture model with hyper-parameter optimization (DAGMM-HO). In our method, the DAGMM-HO applies the conventional DAGMM to the audio domain for the first time, with the idea that its total optimization on reduction of dimensions and statistical modelling will improve the anomaly-detection performance. In addition, the DAGMM-HO solves the hyper-parameter sensitivity problem of the conventional DAGMM by performing hyper-parameter optimization based on the gap statistic and the cumulative eigenvalues. Our evaluation of the proposed method with experimental data of the industrial fans showed that it significantly outperforms previous approaches and achieves up to a 20% improvement based on the standard AUC score.