About this document

Aero-engine Blade Anomaly Detection by Quỳnh Hoàng is a document available to read on EtoBox.

The paper introduces the Aero-engine Blade Anomaly Detection (AeBAD) dataset, which addresses the limitations of existing industrial anomaly detection (IAD) datasets by incorporating domain shifts and varying scales in the data. It also proposes a novel method called masked multi-scale reconstruction (MMR) to enhance anomaly detection performance under these conditions. The findings indicate that MMR outperforms state-of-the-art methods on the AeBAD dataset and achieves competitive results on other datasets

Author

Quỳnh Hoàng

Language

EN