Pantograph unsupervised anomaly detection network based on reverse distillation
Open Access
- 18 October 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 55-60
Abstract
The pantograph, as a crucial component for the transmission of electrical energy in trains, plays a pivotal role in ensuring the safety of train operations. In this study, we have developed an unsupervised anomaly detection network, denoted as Pantograph Anomaly Detection Network (PADN), specifically tailored for the anomaly detection of the pantograph. PADN leverages a reverse distillation architecture and incorporates structural characteristics of the pantograph through a two-phase training approach. Meanwhile, The approach utilizes a trainable dual-teacher architecture, facilitating the assimilation of a more comprehensive range of features specific to the pantograph dataset by the student network. Empirical results demonstrate that PADN surpasses the performance of the current state-of-the-art (SOTA) networks on the pantograph dataset, achieving an accuracy rate of 97.87% in the final identification of pantograph anomalies.Keywords
Funding Information
- Sichuan Provincial Science and Technology Support Program (2023NSFSC0463, and 2023YFG0079)
This publication has 22 references indexed in Scilit:
- Anomaly Detection via Reverse Distillation from One-Class Embedding2022
- Notice of Removal: Multiresolution Knowledge Distillation for Anomaly Detection2021
- PaDiM: A Patch Distribution Modeling Framework for Anomaly Detection and Localization2021
- Novel Vision-Based Abnormal Behavior Localization of Pantograph-Catenary for High-Speed TrainsIEEE Access, 2019
- MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection2019
- OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations2019
- f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networksMedical Image Analysis, 2019
- Adversarially Learned One-Class Classifier for Novelty Detection2018
- Deep Residual Learning for Image Recognition2016
- ImageNet: A large-scale hierarchical image database2009