Regional-Local Adversarially Learned One-Class Classifier Anomalous Sound Detection in Global Long-Term Space

TL;DR

This paper introduces a novel adversarial learning framework for anomalous sound detection that leverages regional and local pattern reconstruction, along with a global filter layer, achieving superior results on industrial datasets.

Contribution

It proposes a multi-pattern adversarial one-class classification method with a global filter layer for improved long-term sound anomaly detection.

Findings

Outperforms recent state-of-the-art ASD methods on four industrial datasets.

Effectively distinguishes normal and abnormal acoustic patterns.

Utilizes a global filter layer for long-term frequency domain interactions.

Abstract

Anomalous sound detection (ASD) is one of the most significant tasks of mechanical equipment monitoring and maintaining in complex industrial systems. In practice, it is vital to precisely identify abnormal status of the working mechanical system, which can further facilitate the failure troubleshooting. In this paper, we propose a multi-pattern adversarial learning one-class classification framework, which allows us to use both the generator and the discriminator of an adversarial model for efficient ASD. The core idea is learning to reconstruct the normal patterns of acoustic data through two different patterns of auto-encoding generators, which succeeds in extending the fundamental role of a discriminator from identifying real and fake data to distinguishing between regional and local pattern reconstructions. Furthermore, we present a global filter layer for long-term interactions in the frequency domain space, which directly learns from the original data without introducing any human priors. Extensive experiments performed on four real-world datasets from different industrial domains (three cavitation datasets provided by SAMSON AG, and one existing publicly) for anomaly detection show superior results, and outperform recent state-of-the-art ASD methods.