Gap Detection of Switch Machines in Complex Environment Based on Object Detection and Image Processing
Publication: Journal of Transportation Engineering, Part A: Systems
Volume 146, Issue 8
Abstract
A turnout, a device to guide tracks, is critical to the safety of high-speed railways. Detecting gaps in switch machines can monitor a turnout’s working performance. Existing gap-detection systems, however, can barely perform at high accuracy and with a low false alarm rate for a long time due to the complex operating conditions of switch machines. This study proposes an approach combining YOLO-based object detection architecture with image processing algorithms, of which YOLO is a deep learning network for object detection. First, YOLO detects target areas in gap images, and then image-processing algorithms identify gaps and calculate gap sizes. This approach targets various types of switch machines and particularly complicated situations. Experiments on gap images of S700K switch machines demonstrate that the accuracy of detecting gaps is 100%, and the accuracy of calculating gap sizes is higher than 99%. Additionally, the proposed approach can exhibit the same high performance on complex images, like overexposed and tilted ones.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request. Data includes gap images; code includes some .py files written in Python; and the models are .dll files.
Acknowledgments
This research is supported by the National Natural Science Foundation of China (61703308), the National Key R&D Program of China (2016YFB1200402), the Fundamental Research Funds for the Central Universities, and Sichuan Science and Technology Program (2019YFG0040). The authors are grateful for the reviewers’ helpful comments and suggestions.
References
Ebadi, M., M. Bagheri, M. S. Lajevardi, and B. Haas. 2019. “Defect detection of railway turnout using 3D scanning.” In Sustainable rail transport, 1–18. Cham, Switzerland: Springer.
Guo, Z. J., H. Ye, W. Dong, Y. W. Yan, and X. Yan. 2017. “A hybrid feature extraction method for fault detection of turnouts.” In Proc., 2017 Chinese Automation Congress, 540–545. New York: IEEE.
Huang, S. Z., X. L. Yang, L. Wang, W. Chen, F. Zhang, and D. C. Dong. 2018. “Two-stage turnout fault diagnosis based on similarity function and fuzzy c-means.” Adv. Mech. Eng. 10 (12): 1–12. https://doi.org/10.1177/1687814018811402.
Huang, S. Z., F. Zhang, R. J. Yu, W. Chen, F. Hu, and D. C. Dong. 2017. “Turnout fault diagnosis through dynamic time warping and signal normalization.” J. Adv. Transp. 2017 (2): 1–8. https://doi.org/10.1155/2017/3192967.
Li, C., L. H. Zhao, and W. N. Liu. 2018. “Automatic detection algorithm of switch machine gap based on canny operator.” J. China Railway Soc. 40 (10): 81–87.
Li, Q. F., C. Li, and J. F. Shi. 2009. “Application of improved Hough transform in the real-time monitoring of notch revealed by switch machine.” In Proc., Int. Symp. on Computer Network and Multimedia Technology, 1–4. Reston, VA: ASCE. https://doi.org/10.1109/CNMT.2009.5374679.
Liu, X., A. Lovett, T. Dick, M. R. Saat, and C. P. L. Barkan. 2014. “Optimization of ultrasonic rail-defect inspection for improving railway transportation safety and efficiency.” J. Transp. Eng. Part A Syst. 140 (10): 04014048. https://doi.org/10.1061/(ASCE)TE.1943-5436.0000697.
Long, B., C. Y. Shang, and D. Zhang. 2013. “Research on switch machine gap monitoring system based on video monitoring technology.” [In Chinese.] Railway Comput. Appl. 22 (11): 43–46.
Redmon, J., S. Divvala, R. Girshick, and A. Farhadi. 2016. “You only look once: Unified, real-time object detection.” In Proc., IEEE Conf. on Computer Vision and Pattern Recognition, 779–788. New York: IEEE.
Xu, T. H., G. Wang, H. F. Wang, T. M. Yuan, and Z. W. Zhong. 2016. “Gap measurement of point machine using adaptive wavelet threshold and mathematical morphology.” Sensors 16 (12): 2006–2017. https://doi.org/10.3390/s16122006.
Xu, W. T., J. P. Zhou, L. C. Yang, and L. Li. 2018. “The implications of high-speed rail for Chinese cities: Connectivity and accessibility.” Transp. Res. Part A Policy Pract. 116 (Oct): 308–326.
Zhang, K. 2014. “The railway turnout fault diagnosis algorithm based on BP neural network.” In Proc., IEEE Int. Conf. on Control Science and Systems Engineering, 135–138. New York: IEEE.
Zhang, K., K. Du, and Y. F. Ju. 2014. “Algorithm of railway turnout fault detection based on PNN neural network.” In Proc., 7th Int. Symp. on Computational Intelligence and Design, 544–547. New York: IEEE. https://doi.org/10.1109/ISCID.2014.140.
Zhang, L., D. C. Dong, J. T. Lin, Z. P. Yuan, and J. Y. Mao. 2012. “A novel method of switch machine gap monitoring based on fiber Bragg grating.” In Vol. 209 of Applied mechanics and materials, 2121–2126. Zürich, Switzerland: Trans Tech.
Zhang, Y. N., and Q. Xie. 2015. “Railway turnout detection method based on machine vision.” [In Chinese.] Comput. Appl. Software 32 (1): 225–228.
Zhong, Z. W., H. B. Zhao, L. F. Zhu, and Q. Y. Xu. 2016. “Detection of switch gaps based on CMOS plane in high-speed railways.” In Proc., Int. Conf. on Progress in Informatics and Computing (PIC), 385–389. New York: IEEE. https://doi.org/10.1109/PIC.2016.7949531.
Zhou, H. B., Y. G. Zhang, L. N. Wang, and C. Cheng. 2014. “Research on machine vision based G detection method of switch machine.” [In Chinese.] Autom. Instrum. 2014 (06): 26–29.
Information & Authors
Information
Published In
Journal of Transportation Engineering, Part A: Systems
Volume 146 • Issue 8 • August 2020
Copyright
©2020 American Society of Civil Engineers.
History
Received: Dec 12, 2019
Accepted: Mar 17, 2020
Published online: Jun 11, 2020
Published in print: Aug 1, 2020
Discussion open until: Nov 11, 2020
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.