A Survey on Audio-Video Based Defect Detection Through Deep Learning in Railway Maintenance
Обзор обнаружения дефектов на основе аудио- и видеоданных с использованием глубокого обучения в техническом обслуживании железнодорожного транспорта
2022-01-01
SCID: 54.1/ke44ybse
Discuss with AI
audio-visual sensorsdeep learningdefect detectionpredictive maintenancerailway maintenance
Figures from the paper
Abstract (AI)
Within Artificial Intelligence, Deep Learning (DL) represents a paradigm that has been showing unprecedented performance in image and audio processing by supporting or even replacing humans in defect and anomaly detection. The Railway sector is expected to benefit from DL applications, especially in predictive maintenance applications, where smart audio and video sensors can be leveraged yet kept distinct from safety-critical functions. Such separation is crucial, as it allows for improving system dependability with no impact on its safety certification. This is further supported by the development of DL in other transportation domains, such as automotive and avionics, opening for knowledge transfer opportunities and highlighting the potential of such a paradigm in railways. In order to summarize the recent state-of-the-art while inquiring about future opportunities, this paper reviews DL approaches for the analysis of data generated by acoustic and visual sensors in railway maintenance applications that have been published until August 31st, 2021. In this paper, the current state of the research is investigated and evaluated using a structured and systematic method, in order to highlight promising approaches and successful applications, as well as to identify available datasets, current limitations, open issues, challenges, and recommendations about future research directions.
Key Findings
1
Audio-video deep learning shows promise for railway defect and anomaly detection, supporting or potentially replacing human inspection in predictive maintenance.
2
Knowledge transfer from deep-learning applications in automotive and avionics provides opportunities for advancing railway maintenance systems.
3
Separating smart audio-video sensing from safety-critical railway functions could improve dependability without affecting safety certification.
4
The paper systematically reviews deep-learning approaches for acoustic and visual sensor data in railway maintenance published through August 31, 2021.
5
The review identifies promising approaches, successful applications, available datasets, limitations, open challenges, and recommended future research directions.
Research Object
audio and video sensor data in railway maintenance applications
Research Subject
deep-learning-based defect and anomaly detection approaches, applications, datasets, limitations, and future research opportunities
Publication Details
Publication Date
2022-01-01
Journal
Publisher
ISSN
Cited by
48
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai9
Deep Residual Learning for Image Recognition2016
ImageNet classification with deep convolutional neural networks2017
Very Deep Convolutional Networks for Large-Scale Image Recognition2014
ImageNet: A large-scale hierarchical image database2009
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network2017
Least Squares Generative Adversarial Networks2017
Deep Learning for Generic Object Detection: A Survey2019
Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network2017
A literature review of Artificial Intelligence applications in railway systems2022