Automated Extraction and Accuracy Verification of Railway Track Axes from MLS Point Clouds
DOI:
https://doi.org/10.55779/ng62635Keywords:
extraction, MLS, point cloud, railway, track axisAbstract
The precise determination of railway track geometry is essential for safe infrastructure operations, yet traditional surveying methods remain operationally intensive. Mobile laser scanning offers a high-efficiency alternative, though utilizing its full potential requires automated processing workflows to convert raw data into vectorized information. This study presented a fully automated methodology for extracting three-dimensional track axes from mobile laser scanning point clouds. The research material consisted of a five-kilometre double-track section captured by a specialized measuring train simultaneously equipped with a mobile laser scanning system and a dedicated track geometry measurement unit. This integrated configuration allowed for the concurrent acquisition of point cloud data and a high-precision reference trajectory within an identical coordinate frame, effectively eliminating external co-registration errors. The proposed algorithm utilized temporal segmentation, Euclidean clustering, and geometric filtering to identify rail head candidates and derive the axis geometry without manual intervention. The results demonstrated high geometric fidelity, with the mean absolute error remaining below 10 mm across all spatial dimensions. Specifically, the three-dimensional spatial mean absolute error was 5.9 mm for the left track and 9.1 mm for the right track. Detailed spatial analysis revealed that the largest deviations were concentrated in turnout sections and superelevated curves. Furthermore, the investigation identified systematic horizontal biases attributable to inaccuracies in the reference dataset rather than the extraction algorithm. The findings confirmed that mobile laser scanning provides a viable, efficient alternative to static surveying for large-scale network monitoring.
Metrics
References
Ariyachandra M, Wen Y, Yu J (2025). Advancing rail infrastructure: Integrating digital twins and cyber-physical systems for predictive maintenance. Proceedings of the European Conference on Computing in Construction. https://doi.org/10.35490/EC3.2025.272
Chen Q, Niu X, Zuo L, Zhang T, Xiao F, Liu Y, et al. (2018). A railway track geometry measuring trolley system based on aided INS. Sensors 18(2): 538. https://doi.org/10.3390/s18020538
Elberink SO, Khoshelham K (2015). Automatic extraction of railroad centerlines from mobile laser scanning data. Remote Sensing 7(5): 5565–5583. https://doi.org/10.3390/rs70505565
Irlik M (2017). Linear positioning of railway objects. Scientific Journal of Silesian University of Technology. Series Transport 96: 49–57. https://doi.org/10.20858/sjsutst.2017.96.5
Izvoltova J, Villim A, Kozak P (2014). Determination of geometrical track position by robotic total station. Procedia Engineering 91: 322–327. https://doi.org/10.1016/j.proeng.2014.12.068
Judek S, Wilk A, Koc W, Lewiński L, Szumisz A, Chrostowski P, et al. (2022). Preparatory railway track geometry estimation based on GNSS and IMU systems. Remote Sensing 14(21): 5472. https://doi.org/10.3390/rs14215472
Koehl M, Schueller D, Barrot O, Guillemin S (2022). 3D modelling of tram tunnel from TLS point clouds for technical documentation and extraction of characteristic lines. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLIII-B2-2022: 1047–1054. https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-1047-2022
Kregar K, Možina J, Ambrožič T, Kogoj D, Marjetič A, Štebe G, et al. (2017). Control measurements of crane rails performed by terrestrial laser scanning. Sensors 17(7): 1671. https://doi.org/10.3390/s17071671
Marais J, Beugin J, Berbineau M (2017). A survey of GNSS-based research and developments for the European railway signaling. IEEE Transactions on Intelligent Transportation Systems 18(10): 2602–2618. https://doi.org/10.1109/TITS.2017.2658179
Mikhaylov D, Amatetti C, Polonelli T, Masina E, Campana R, Berszin K, et al. (2023). Toward the future generation of railway localization exploiting RTK and GNSS. IEEE Transactions on Instrumentation and Measurement 72: 1–12. https://doi.org/10.1109/TIM.2023.3272048
Niina Y, Honma R, Honma Y, Kondo K, Tsuji K, Hiramatsu T, et al. (2018). Automatic rail extraction and clearance check with a point cloud captured by MLS in a railway. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2: 767–771. https://doi.org/10.5194/isprs-archives-XLII-2-767-2018
Rashdi R, Garrido I, Balado J, Del Río-Barral P, Rodríguez-Somoza JL, Martínez-Sánchez J (2024). Comparative evaluation of LiDAR systems for transport infrastructure: Case studies and performance analysis. European Journal of Remote Sensing 57(1): 2316304. https://doi.org/10.1080/22797254.2024.2316304
Rusu RB, Cousins S (2011). 3D is here: Point cloud library (PCL). Proceedings of the IEEE International Conference on Robotics and Automation: 1–4. https://doi.org/10.1109/ICRA.2011.5980567
Sala AJ, Felez J, Cano-Moreno JD (2023). Efficient railway turnout design: Leveraging TRIZ-based approaches. Applied Sciences 13(17): 9531. https://doi.org/10.3390/app13179531
Shankar S, Roth M, Schubert LA, Verstegen JA (2020). Automatic mapping of center line of railway tracks using global navigation satellite system, inertial measurement unit and laser scanner. Remote Sensing 12(3): 411. https://doi.org/10.3390/rs12030411
Soilán M, Nóvoa A, Sánchez-Rodríguez A, Justo A, Riveiro B (2021). Fully automated methodology for the delineation of railway lanes and the generation of IFC alignment models using 3D point cloud data. Automation in Construction 126: 103684. https://doi.org/10.1016/j.autcon.2021.103684
Soni A, Robson S, Gleeson B (2014). Extracting rail track geometry from static terrestrial laser scans for monitoring purposes. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XL-5: 553–557. https://doi.org/10.5194/isprsarchives-XL-5-553-2014
Specht C, Wilk A, Koc W, Karwowski K, Dąbrowski P, Specht M, et al. (2020). Verification of GNSS measurements of the railway track using standard techniques for determining coordinates. Remote Sensing 12(18): 2874. https://doi.org/10.3390/rs12182874
Stallo C, Neri A, Salvatori P, Capua R, Rispoli F (2019). GNSS integrity monitoring for rail applications: Two-tiers method. IEEE Transactions on Aerospace and Electronic Systems 55(4): 1850–1863. https://doi.org/10.1109/TAES.2018.2876735
Stein D (2017). Mobile laser scanning based determination of railway network topology and branching direction on turnouts. Karlsruher Institut für Technologie.
Ton B, Akster R (2025). Large scale asset detection within railway scene point cloud data from mobile laser scanning. IEEE Access 13: 129114–129126. https://doi.org/10.1109/ACCESS.2025.3590779
Wilk A, Koc W, Specht C, Skibicki J, Judek S, Karwowski K, et al. (2021). Innovative mobile method to determine railway track axis position in global coordinate system using position measurements performed with GNSS and fixed base of the measuring vehicle. Measurement 175: 109016. https://doi.org/10.1016/j.measurement.2021.109016
Yang B, Fang L (2014). Automated extraction of 3D railway tracks from mobile laser scanning point clouds. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 7(12): 4750–4761. https://doi.org/10.1109/JSTARS.2014.2312378
Yao L, Zhang S, Wang Z, Sun H, Chen Q, Gilbert KM (2020). Metro gauge inspection system based on mobile laser scanning technology. Survey Review 52(375): 531–543. https://doi.org/10.1080/00396265.2019.1661164
Zampogiannis K, Fermüller C, Aloimonos Y (2018). Cilantro: A lean, versatile, and efficient library for point cloud data processing. Proceedings of the ACM International Conference on Multimedia: 1364–1367. https://doi.org/10.1145/3240508.3243655
Zhang R, Ding Q, Ng AHM, Wang D, Deng J, Xu M, et al. (2025). TripletA-Net: A deep learning model for automatic railway track extraction from airborne LiDAR point clouds. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 18: 9195–9210. https://doi.org/10.1109/JSTARS.2025.3555292
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Zoltán Nagy, Árpád Somogyi
This work is licensed under a Creative Commons Attribution 4.0 International License.
Open Access, Copyright, and Licensing
All content published in Nova Geodesia is made immediately available on an open-access basis under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
Authors retain copyright in their published work. By submitting and publishing an article in the journal, the authors grant Nova Geodesia and its publisher, the Society of Land Measurements and Cadastre from Transylvania (SMTCT), a non-exclusive right to publish, distribute, reproduce, archive, preserve, and disseminate the article in all formats and media.
Under the CC BY 4.0 license, users may read, download, copy, distribute, print, search, link to, adapt, and otherwise reuse the published content for any lawful purpose, including commercial use, provided that appropriate credit is given to the authors and the original publication, a link to the license is provided, and any changes are indicated.
The full terms of the license are available at:
https://creativecommons.org/licenses/by/4.0/
Authors are responsible for obtaining permission to reproduce any third-party material that is not covered by the article’s CC BY 4.0 license.
No submission fee, article processing charge, publication fee, or other mandatory publication charge is imposed by the journal.
