Automatic detection of arbitrarily oriented fastener defect in high-speed railway

Abstract In high-speed railways, contact forces between the pantograph and the overhead catenary system (OCS) are substantial. Vibration and excitation propagated from the vehicle-track and the pantograph-OCS interactions would progressively damage fasteners, including but not limited to, loose bolts, cracked components, and missing parts. Existing automatic detection methods typically rely on a three-stage approach, of which the first two stages focus on locating joints and fasteners while the last stage focuses on the detection. Due to the nature of the three-stage detector, the computational cost is high, and the inspection speed is low. This study proposes an innovative two-stage method with two improved convolutional neural network (CNN)-based networks, cascade YOLO (You Only Look Once) and Rotation RetinaNet (RRNet). The proposed method was compared to traditional horizontal anchor-based methods and other methods. The results demonstrate the proposed method outperforms other methods in terms of accuracy, while maintaining a reasonably high processing speed.

[1]  Yang Lei,et al.  Railway Subgrade Defect Automatic Recognition Method Based on Improved Faster R-CNN , 2018, Sci. Program..

[2]  Ye Han,et al.  A CNN-Based Defect Inspection Method for Catenary Split Pins in High-Speed Railway , 2019, IEEE Transactions on Instrumentation and Measurement.

[3]  Yong Qin,et al.  A UAV-Based Visual Inspection Method for Rail Surface Defects , 2018, Applied Sciences.

[4]  Feng Guo,et al.  Automatic railroad track components inspection using real‐time instance segmentation , 2020, Comput. Aided Civ. Infrastructure Eng..

[5]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[6]  Yunyi Jia,et al.  Automatic detection of moisture damages in asphalt pavements from GPR data with deep CNN and IRS method , 2020 .

[7]  Ali Daadbin,et al.  Development, Testing and Implementation of the Pantograph Damage Assessment System (PANDAS) , 2010 .

[8]  Xiaoping Ma,et al.  Densely pyramidal residual network for UAV-based railway images dehazing , 2020, Neurocomputing.

[9]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[10]  Mehmet Karaköse,et al.  Anomaly detection using a modified kernel-based tracking in the pantograph-catenary system , 2015, Expert Syst. Appl..

[11]  Shubin Li,et al.  Detection of rail surface defects based on CNN image recognition and classification , 2018, 2018 20th International Conference on Advanced Communication Technology (ICACT).

[12]  Tianxu Zhang,et al.  Joint horizontal-vertical enhancement and tracking scheme for robust contact-point detection from pantograph-catenary infrared images , 2020 .

[13]  Xiangyang Xue,et al.  Arbitrary-Oriented Scene Text Detection via Rotation Proposals , 2017, IEEE Transactions on Multimedia.

[14]  Zhigang Liu,et al.  High-speed railway catenary components detection using the cascaded convolutional neural networks , 2017, 2017 IEEE International Conference on Imaging Systems and Techniques (IST).

[15]  P. Boffi,et al.  Optical Fiber Sensors to Measure Collector Performance in the Pantograph-Catenary Interaction , 2009, IEEE Sensors Journal.

[16]  Kai Liu,et al.  Deep learning-based visual ensemble method for high-speed railway catenary clevis fracture detection , 2020, Neurocomputing.

[17]  Yu Qian,et al.  Real-time railroad track components inspection based on the improved YOLOv4 framework , 2021 .

[18]  Zhigang Liu,et al.  A High-Precision Loose Strands Diagnosis Approach for Isoelectric Line in High-Speed Railway , 2018, IEEE Transactions on Industrial Informatics.

[19]  Richard O. Duda,et al.  Use of the Hough transformation to detect lines and curves in pictures , 1972, CACM.

[20]  Yang Yang,et al.  Instance-level recognition and quantification for concrete surface bughole based on deep learning , 2019, Automation in Construction.

[21]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Peng Dai,et al.  Rail fastener automatic recognition method in complex background , 2018, International Conference on Digital Image Processing.

[23]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[24]  Yong Qin,et al.  Defect Detection for Catenary Sling Based on Image Processing and Deep Learning Method , 2019 .

[25]  Jian Li,et al.  Image Registration-Based Bolt Loosening Detection of Steel Joints , 2018, Sensors.

[26]  Jian Li,et al.  Non-contact fatigue crack detection in civil infrastructure through image overlapping and crack breathing sensing , 2019, Automation in Construction.

[27]  Mitsuru Ikeda,et al.  Method for Detecting Step-Shaped Wear on Contact Strips by Measuring Catenary Vibration , 2011 .

[28]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[29]  Luhua Fu,et al.  Automatic measuring method of catenary geometric parameters based on laser scanning and imaging , 2018, International Conference on Optical Instruments and Technology.

[30]  Zhang Yi,et al.  An improved tiny-yolov3 pedestrian detection algorithm , 2019, Optik.

[31]  D Marr,et al.  Theory of edge detection , 1979, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[32]  Elzbieta Pastucha,et al.  Catenary System Detection, Localization and Classification Using Mobile Scanning Data , 2016, Remote Sensing.

[33]  Zhi Suo,et al.  Automatic Rail Surface Defects Inspection Based on Mask R-CNN , 2021, Transportation Research Record: Journal of the Transportation Research Board.

[34]  Zhigang Liu,et al.  Automatic recognition for catenary insulators of high-speed railway based on contourlet transform and Chan–Vese model , 2016 .

[35]  Shuicheng Yan,et al.  An HOG-LBP human detector with partial occlusion handling , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[36]  Peter E.D. Love,et al.  Convolutional neural networks: Computer vision-based workforce activity assessment in construction , 2018, Automation in Construction.

[37]  Shuai Li,et al.  Integrated Processing of Image and GPR Data for Automated Pothole Detection , 2016, J. Comput. Civ. Eng..

[38]  Xinzheng Lu,et al.  Automated regional seismic damage assessment of buildings using an unmanned aerial vehicle and a convolutional neural network , 2020 .

[39]  Mitsuru Ikeda,et al.  The results of the pantograph–catenary interaction benchmark , 2015 .

[40]  François Clemens,et al.  A defect classification methodology for sewer image sets with convolutional neural networks , 2019, Automation in Construction.

[41]  Yong Qin,et al.  Hybrid deep learning architecture for rail surface segmentation and surface defect detection , 2021, Comput. Aided Civ. Infrastructure Eng..

[42]  Young-Jin Cha,et al.  Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning , 2020 .

[43]  Shibin Gao,et al.  Deep Architecture for High-Speed Railway Insulator Surface Defect Detection: Denoising Autoencoder With Multitask Learning , 2019, IEEE Transactions on Instrumentation and Measurement.

[44]  Mani Golparvar-Fard,et al.  Human-object interaction recognition for automatic construction site safety inspection , 2020 .

[45]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[46]  Peng Yang,et al.  Railway obstacle detection algorithm using neural network , 2018 .

[47]  Zhigang Liu,et al.  Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network , 2018, IEEE Transactions on Instrumentation and Measurement.