Comparative Analysis of Lightweight CNN Architectures for Railway Track Fault Detection
Main Article Content
Abstract
The application of artificial intelligence is widely used in various industrial sectors, including the transportation sector, one of which is the use of Image Processing to detect damage to railway lines. Railway conventional inspections involve visually examining and measuring railway infrastructure to identify potential problems. These inspections are an important aspect of ensuring the safety and efficiency of the railway network. In some places, railway track inspections still use conventional methods with electricity and vision. The use of artificial intelligence is expected to minimize errors, increase efficiency, reduce time and costs in train damage inspections. This research aims to find the best architecture and its development is expected to be used Accelerate the process of locating damage so that railroads can be repaired immediately. In this study, we evaluate the CNN model with the lightweight model to classify the image of the condition of the train track. Several types of lightweight models chosen are EfficientNetB0, EfficientNetB3, MobileNetV2, NasNetMobile. From the results of the evaluation carried out, it was found that EfficientNetB0 was 0.875, EfficientNetB3 was 0.958, MobileNetV2 was 0.917, and NasNetMobile was 0.8333.
Article Details
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
References
References
C. Tastimur, M. Karakose, and E. Akin, “Image Processing Based Level Crossing Detection and Foreign Objects Recognition Approach in Railways,” International Journal of Applied Mathematics Electronics and Computers, no. Special Issue-1, pp. 19–23, Sep. 2017.
J. Jang, M. Shin, S. Lim, J. Park, J. Kim, and J. Paik, “Intelligent Image-Based Railway Inspection System Using Deep Learning-Based Object Detection and Weber Contrast-Based Image Comparison,” Sensors, vol. 19, no. 21, p. 4738, Oct. 2019.
Y. Rubinsztejn and K. Chen, “Automatic Detection of Objects of Interest from Rail Track Images,” M.S. thesis, School of Computer Science, University of Manchester, Manchester, U.K., 2011
E. Deutschl, C. Gasser, A. Niel, and J. Werschonig, “Defect Detection on Rail Surfaces by a Vision Based System,” in Proceedings of the IEEE Intelligent Vehicles Symposium, Parma, Italy, Jun. 2004, pp. 507–510.
M. Deutschl, T. Heiler, and S. Schmid, “Defect Detection on Rail Surfaces by a Vision Based System,” in Proceedings of the 2004 IEEE International Conference on Systems, Man and Cybernetics, vol. 3, pp. 2271–2276, 2004.
M. Al Rabbani dan M. Hussain, "Lightweight Convolutional Network with Integrated Attention Mechanism for Missing Bolt Detection in Railways," Metrology, vol. 4, no. 2, pp. 254–278, 2024.
Z. Zhao, J. Kang, Z. Sun, T. Ye, dan B. Wu, "A Real-Time and High-Accuracy Railway Obstacle Detection Method Using Lightweight CNN and Improved Transformer," Measurement, vol. 229, p. 115380, 2024.
J. Li, Y. Fu, D. Yan, S. L. Ma, and C.-W. Sham, "An Edge AI System Based on FPGA Platform for Railway Fault Detection," in Proceedings of the 2024 IEEE 13th Global Conference on Consumer Electronics (GCCE), Auckland, New Zealand, Oct. 2024, pp. 1387–1389.
Z. Zheng, Y. Zhang, and H. Li, "A Real-Time Railway Fastener Inspection Method Using the Lightweight Depth Estimation Network," Measurement, vol. 183, p. 109780, 2021.
T. Wang, Z. Zhang, F. Yang, and K. L. Tsui, "Automatic Rail Component Detection Based on AttnConv-Net," IEEE Sensors Journal, vol. 22, no. 3, pp. 2379–2388, 2022.
M. Tan and Q. V. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks," in Proceedings of the 36th International Conference on Machine Learning (ICML), Long Beach, CA, USA, 2019, pp. 6105–6114.
M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "MobileNetV2: Inverted Residuals and Linear Bottlenecks," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 2018, pp. 4510–4520.
B. Zoph, V. Vasudevan, J. Shlens, and Q. V. Le, "Learning Transferable Architectures for Scalable Image Recognition," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 2018, pp. 8697–8710.
A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017.