elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Rail Surface Defect Detection and Severity Analysis Using CNNs on Camera and Axle Box Acceleration Data

Jahan, Kanwal and Lähns, Alexander and Baasch, Benjamin and Heusel, Judith and Roth, Michael (2024) Rail Surface Defect Detection and Severity Analysis Using CNNs on Camera and Axle Box Acceleration Data. In: Lecture Notes in Mechanical Engineering, pp. 423-435. Springer. 7th International Congress and Workshop on Industrial AI and eMaintenance, IAI 2023, 2023-06-13 - 2023-06-15, Luleå, Schweden. doi: 10.1007/978-3-031-39619-9_31. ISBN 978-303038076-2. ISSN 2195-4356.

[img] PDF - Only accessible within DLR until 1 January 2025
1MB

Official URL: https://link.springer.com/chapter/10.1007/978-3-031-39619-9_31

Abstract

Rail surface defect detection is a relevant problem in the field of data-driven railway maintenance. Artificial intelligence and neural networks (NN) for axle box acceleration (ABA) or camera data show great potential for defect detection and classification. However, a sufficient amount of labeled training data is required, all the more if the defect severity is to be estimated. A unique dataset of time-synchronized ABA and camera data is employed that contains labeled defect instances. For the image analysis, RetinaNet as a single-stage object detector (with the backbone of ResNet-50 and a feature pyramid network) is used to achieve high classification performance for the two most common rail surface defects (squat and corrugation). Additionally, a machine learning-based method on ABA data to estimate defect severity levels (low, medium, heavy) is proposed. False positives are detected in the original labels by both classifiers during evaluation. The inspection of the false positives in image data reveals that defects have been overlooked in the initial labeling. The insights of this work help to reduce the dependency on labeled data by using only a few labeled samples and by exploiting complementary data sources instead of increasing the number of labeled instances.

Item URL in elib:https://elib.dlr.de/201722/
Document Type:Conference or Workshop Item (Speech)
Title:Rail Surface Defect Detection and Severity Analysis Using CNNs on Camera and Axle Box Acceleration Data
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Jahan, KanwalUNSPECIFIEDhttps://orcid.org/0009-0000-6977-239XUNSPECIFIED
Lähns, AlexanderUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Baasch, BenjaminUNSPECIFIEDhttps://orcid.org/0000-0003-1970-3964UNSPECIFIED
Heusel, JudithUNSPECIFIEDhttps://orcid.org/0009-0007-7573-6652UNSPECIFIED
Roth, MichaelUNSPECIFIEDhttps://orcid.org/0000-0002-4812-346XUNSPECIFIED
Date:1 January 2024
Journal or Publication Title:Lecture Notes in Mechanical Engineering
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:No
DOI:10.1007/978-3-031-39619-9_31
Page Range:pp. 423-435
Publisher:Springer
Series Name:Lecture Notes in Mechanical Engineering
ISSN:2195-4356
ISBN:978-303038076-2
Status:Published
Keywords:Deep learning, Convolutional neural network, Time-synchronized dataset, Supervised learning, Severity analysis, Rail surface defects Squats, Corrugation
Event Title:7th International Congress and Workshop on Industrial AI and eMaintenance, IAI 2023
Event Location:Luleå, Schweden
Event Type:international Conference
Event Start Date:13 June 2023
Event End Date:15 June 2023
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Rail Transport
DLR - Research area:Transport
DLR - Program:V SC Schienenverkehr
DLR - Research theme (Project):V - TraCo - Train Control and Management
Location: Berlin-Adlershof , Braunschweig
Institutes and Institutions:Institute of Transportation Systems > Information Gathering and Modelling, BA
Institute of Transportation Systems > Information Gathering and Modelling, BS
Deposited By: Baasch, Dr. Benjamin
Deposited On:29 Jan 2024 09:56
Last Modified:24 Apr 2024 21:02

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.