Corrugation and Squat Classification and Detection with VGG16 and YOLOv5 Neural Network Models

Muhammad Syukri Mohd Yazed - Universiti Tun Hussein Onn Malaysia, Johor, Malaysia
Mohd Amin Mohd Yunus - Universiti Tun Hussein Onn Malaysia, Johor, Malaysia
Ezak Fadzrin Ahmad Shaubari - Universiti Tun Hussein Onn Malaysia, Johor, Malaysia
Nor Aziati Abdul Hamid - Universiti Tun Hussein Onn Malaysia, Johor, Malaysia
Azmale Amzah - Track Network Maintenance Ampang Line, Selangor, 68000, Malaysia
Zulhelmi Md Ali - Track Network Maintenance Ampang Line, Selangor, 68000, Malaysia


Citation Format:



DOI: http://dx.doi.org/10.62527/joiv.8.2.2756

Abstract


Railway track defects in Malaysia pose significant risks of train derailments and accidents, underscoring the urgency for early and accurate defect detection and classification. This study presents a novel approach utilizing deep learning models, VGG16 and YOLOv5, for detecting and classifying railway track defects, explicitly focusing on corrugation and squat defects. The research's uniqueness lies in its application of these specific models and the composition of a dataset collected from extensive field measurements and inspections across various railway tracks within the Track Network Maintenance Ampang Line in Malaysia. The results demonstrate that these models achieve high precision in defect classification and detection of defects by more than 80%. The proposed methodology provides the railway industry with a powerful tool to streamline maintenance planning and prioritize defect remediation efficiently. Early defect detection can prevent potential accidents and improve safety and operational efficiency. Future studies can expand on these findings by exploring the extension of the proposed techniques to address other types of rail defects. Incorporating a diverse range of scenarios and operating conditions in the dataset could further enhance the models' performance and generalization. Real-time deployment and integration with existing maintenance systems are crucial for practical adoption. This research has strengths but acknowledges limitations. Additional evaluation metrics and a diverse dataset are essential for model performance. Leveraging deep learning models offers a reliable solution for railway maintenance, enhancing safety and efficiency. Addressing these limitations will drive proactive defect management, ensuring safe and reliable railway networks.

Keywords


Railways defect, corrugation; squat; neural network; VGG16; YOLOv5

Full Text:

PDF

References


Z. Zakri et al., “Utilizing Rolling Noise for Rail Surface Defects Detection on Wayside Rail Track in Malaysia,” in AIP Conference Proceedings, American Institute of Physics Inc., May 2023. doi: 10.1063/5.0121415.

X. Zhang, X. Li, and H. Zhang, "Railway track defect detection using deep learning," IEEE Access, vol. 9, pp. 3076888, 2021. doi: 10.1109/ACCESS.2021.3076888.

W. Gong, M. F. Akbar, G. N. Jawad, M. F. P. Mohamed, and M. N. A. Wahab, “Nondestructive Testing Technologies for Rail Inspection: A Review,” Coatings, vol. 12, no. 11. MDPI, Nov. 01, 2022. doi: 10.3390/coatings12111790.

J. W. Park et al., “Rail Surface Defect Detection and Analysis Using Multi-Channel Eddy Current Method Based Algorithm for Defect Evaluation,” J Nondestr Eval, vol. 40, no. 3, Sep. 2021, doi: 10.1007/s10921-021-00810-9.

A. Synodinos, “Identification of railway track components and defects by analysis of wheel-rail interaction noise Athens, Greece,” in 23rd International Congress on Sound and Vibration , 2016. [Online]. Available: https://www.researchgate.net/publication/315475757

P. Cawley, “Non-destructive testing - current capabilities and future directions,” Journal of Materials Design and Applications, vol. 215, no. 4, pp. 213–223, 2001, doi: https://doi.org/10.1177/146442070121500403.

J. Wang, J. Li, and Y. Wang, "A novel method for railway track defect detection using a convolutional neural network with attention mechanism," IEEE Access, vol. 9, pp. 3076889, 2021. doi: 10.1109/ACCESS.2021.3076889.

E. Deutschl, C. Gasser, A. Niel, and J. Werschonig, “Defect detection on rail surfaces by a vision based system,” in IEEE Intelligent Vehicles Symposium, Proceedings, 2004, pp. 507–511. doi: 10.1109/ivs.2004.1336435.

X. Zhang, X. Li, H. Zhang, and X. Yang, "A deep learning-based method for railway track defect detection using 3D images," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 1, pp. 310-319, 2022. doi: 10.1109/TITS.2022.3176913.

H. Shi, L. Zhuang, X. Xu, Z. Yu, and L. Zhu, “An ultrasonic guided wave mode selection and excitation method in rail defect detection,” Applied Sciences (Switzerland), vol. 9, no. 6, 2019, doi: 10.3390/app9061170.

F. Wu, X. Xie, J. Guo, and Q. Li, "Internal defects detection method of the railway track based on generalization features cluster under ultrasonic images," Chinese Journal of Mechanical Engineering, vol. 35, no. 1, pp. 59-67, 2022. doi: 10.1186/s10033-022-00726-z.

W. Wang et al., “Novel Coil Transducer Induced Thermoacoustic Detection of Rail Internal Defects towards Intelligent Processing,” IEEE Transactions on Industrial Electronics, 2023, doi: 10.1109/TIE.2023.3250766.

R. Yang, Y. He, B. Gao, G. Yun Tian, and J. Peng, “Lateral heat conduction based eddy current thermography for detection of parallel cracks and rail tread oblique cracks,” Measurement (Lond), vol. 66, pp. 54–61, 2015, doi: 10.1016/j.measurement.2015.01.024.

D. Tabernik, S. Šela, J. Skvarč, and D. Skočaj, “Segmentation-based deep-learning approach for surface-defect detection,” J Intell Manuf, vol. 31, no. 3, pp. 759–776, Mar. 2020, doi: 10.1007/s10845-019-01476-x.

J. Wang, J. Li, and Y. Wang, "A novel method for railway track defect detection using deep learning," Sensors, vol. 19, no. 13, pp. 3432, 2019. doi: 10.3390/s19133432.

X. Zhang, R. Wang, Z. Li, and J. Wang, "Railway track defect detection using a cascaded convolutional neural network," *IEEE Transactions on Intelligent Transportation Systems*, vol. 22, no. 1, pp. 160-169, 2021. doi: 10.1109/TITS.2021.3072927

Z.-J. Wan and S.-Q. Chen, "Railway Tracks Defects Detection Based on Deep Convolution Neural Networks," in Artificial Intelligence in China, Q. Liang, W. Wang, J. Mu, X. Liu, Z. Na, and X. Cai, Eds. Singapore: Springer Singapore, 2021, pp. 119-129. doi: 10.1007/978-981-15-8599-9_15.

Q. Li and S. Ren, “A visual detection system for rail surface defects,” IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 42, no. 6, pp. 1531–1542, 2012, doi: 10.1109/TSMCC.2012.2198814.

Q. Li, Y. Tan, Z. Huayan, S. Ren, P. Dai, and W. Li, “A Visual Inspection System for Rail Corrugation Based on Local Frequency Features,” in 2016 IEEE 14th Intl Conf on Dependable, Autonomic and Secure Computing, 14th Intl Conf on Pervasive Intelligence and Computing, 2nd Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress(DASC/PiCom/DataCom/CyberSciTech), IEEE, Aug. 2016, pp. 18–23. doi: 10.1109/DASC-PICom-DataCom-CyberSciTec.2016.19.

H. Lee, J. Hong, T. W. Wendimagegn, and H. Lee, “Rail corrugation detection and characterization using computer vision,” Sensors, vol. 21, no. 24, Dec. 2021, doi: 10.3390/s21248335.

S. Mittal and D. Rao, “Vision Based Railway Track Monitoring using Deep Learning.” [Online]. Available: https://www.researchgate.net/publication/321160560

X. Wei, Z. Yang, Y Automatic railway track defect detection using deep convolutional neural networks. Liu, D. Wei, L. Jia, and Y. Li, “Railway track fastener defect detection based on image processing and deep learning techniques: A comparative study,” Eng Appl Artif Intell, vol. 80, pp. 66–81, Apr. 2019, doi: 10.1016/j.engappai.2019.01.008.

S. Yanan, Z. Hui, L. Li, and Z. Hang, “Rail Surface Defect Detection Method Based on YOLOv3 Deep Learning Networks,” in Proceedings 2018 Chinese Automation Congress, CAC 2018, Institute of Electrical and Electronics Engineers Inc., Jan. 2019, pp. 1563–1568. doi: 10.1109/CAC.2018.8623082.

X. Wei, D. Wei, D. Suo, L. Jia, and Y. Li, “Multi-target defect identification for railway track line based on image processing and improved YOLOv3 model,” IEEE Access, vol. 8, pp. 61973–61988, 2020, doi: 10.1109/ACCESS.2020.2984264.

A. Bracciali and G. Cascini, “Detection of corrugation and wheelflats of railway wheels using energy and cepstrum analysis of rail acceleration,” Proc. Instn. Mech. Engrs, vol. 211, no. F, pp. 109–116, 1997, doi: 10.1243/0954409971530950.

H. Xu, D. Huber, and E. J. Heller, “Cross section and optical theorem for defects on a corrugated surface,” J Chem Phys, vol. 89, no. 4, pp. 2550–2557, 1988, doi: 10.1063/1.455050.

D. S. Reddy and P. R. C. Prasad, “Prediction of vegetation dynamics using NDVI time series data and LSTM,” Model Earth Syst Environ, vol. 4, no. 1, pp. 409–419, Apr. 2018, doi: 10.1007/s40808-018-0431-3.

E. Bisong, "Google colaboratory," Building machine learning and deep learning models on google cloud platform: a comprehensive guide for beginners, pp. 59-64, Sep. 2019. doi: 10.1007/978-1-4842-4470-8_7.

D. Theckedath and R. R. Sedamkar, “Detecting Affect States Using VGG16, ResNet50 and SE-ResNet50 Networks,” SN Comput Sci, vol. 1, no. 2, Mar. 2020, doi: 10.1007/s42979-020-0114-9

J. Sresakoolchai and S. Kaewunruen, "Railway defect detection based on track geometry using supervised and unsupervised machine learning," Structural Health Monitoring, vol. 21, no. 4, pp. 1757-1767, 2022. doi: 10.1177/14759217211044492.