{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T16:05:03Z","timestamp":1783526703330,"version":"3.55.0"},"reference-count":30,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2021,10,30]],"date-time":"2021-10-30T00:00:00Z","timestamp":1635552000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"General Project of Scientific Research Program of Beijing Municipal Education Commission","award":["KM202010016003"],"award-info":[{"award-number":["KM202010016003"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["51975038"],"award-info":[{"award-number":["51975038"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["51875032"],"award-info":[{"award-number":["51875032"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.<\/jats:p>","DOI":"10.3390\/e23111437","type":"journal-article","created":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T22:21:08Z","timestamp":1635805268000},"page":"1437","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":64,"title":["A Study on Railway Surface Defects Detection Based on Machine Vision"],"prefix":"10.3390","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0841-035X","authenticated-orcid":false,"given":"Tangbo","family":"Bai","sequence":"first","affiliation":[{"name":"School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"},{"name":"Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jialin","family":"Gao","sequence":"additional","affiliation":[{"name":"School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"},{"name":"Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianwei","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"},{"name":"Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dechen","family":"Yao","sequence":"additional","affiliation":[{"name":"School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"},{"name":"Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles, Beijing University of Civil Engineering and Architecture, Beijing 100044, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2021,10,30]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Nenov, N., Dimitrov, E., Vasilev, V., and Piskulev, P. (2011, January 11\u201315). Sensor system of detecting defects in wheels of railway vehicles running at operational speed. Proceedings of the 2011 34th International Spring Seminar on Electronics Technology, Tratanska Lomnica, Slovakia.","DOI":"10.1109\/ISSE.2011.6053953"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"760","DOI":"10.1109\/TITS.2013.2287155","article-title":"Rail component detection, optimization, and assessment for automatic rail track inspection","volume":"15","author":"Li","year":"2014","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1980","DOI":"10.1109\/TITS.2014.2307955","article-title":"Automatic detection of squats in railway infrastructure","volume":"15","author":"Molodova","year":"2014","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_4","first-page":"14","article-title":"Wayside train monitoring systems: Networking for greater safety","volume":"17","author":"Badran","year":"2011","journal-title":"Glob. Railw. Rev."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Ho, T.K., Liu, S.Y., Ho, Y.T., Ho, K.H., Wong, K.K., Lee, K.Y., Tam, H.Y., and Ho, S.L. (2008, January 18\u201320). Signature analysis on wheel-rail interaction for rail defect detection. Proceedings of the 2008 4th IET International Conference on Railway Condition Monitoring, Derby, UK.","DOI":"10.1049\/ic:20080342"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1016\/j.ndteint.2003.06.002","article-title":"Rail flaw detection: Overview and needs for future developments","volume":"37","author":"Clark","year":"2004","journal-title":"NDT E Int."},{"key":"ref_7","first-page":"546","article-title":"Detection of damage and crack in railhead by using eddy current testing","volume":"3","author":"Song","year":"2011","journal-title":"J. Electromagn. Anal. Appl."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"125","DOI":"10.3141\/2448-15","article-title":"Detection of range-based rail gage and missing rail fasteners: Use of high-resolution two- and three-dimensional images","volume":"2448","author":"Lorente","year":"2014","journal-title":"Transp. Res. Rec."},{"key":"ref_9","first-page":"3692","article-title":"Development of an ultra-wideband radar system for vehicle detection at railway crossings","volume":"6","author":"Lohmeier","year":"2002","journal-title":"IEEE Int. Geosci. Remote Sens. Symp."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1016\/S0168-1699(02)00100-X","article-title":"Machine vision technology for agricultural applications","volume":"36","author":"Chen","year":"2002","journal-title":"Comput. Electron. Agric."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"2189","DOI":"10.1109\/TIM.2012.2184959","article-title":"A real-time visual inspection system for discrete surface defects of rail heads","volume":"61","author":"Li","year":"2012","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_12","first-page":"220","article-title":"Rail surface defects detection based on gray scale gradient characteristics of image","volume":"9","author":"Min","year":"2018","journal-title":"Chin. J. Instrum."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1109\/TITS.2016.2521866","article-title":"Bilevel feature extraction-based text mining for fault diagnosis of railway systems","volume":"18","author":"Wang","year":"2017","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"113211","DOI":"10.1016\/j.eswa.2020.113211","article-title":"An Automatic Nucleus Segmentation and CNN model based classification method of white blood cell","volume":"149","author":"Banik","year":"2020","journal-title":"Expert Syst. Appl."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1531","DOI":"10.1109\/TSMCC.2012.2198814","article-title":"A visual detection system for rail surface defects","volume":"42","author":"Li","year":"2012","journal-title":"IEEE Trans. Syst. Man Cybern. Part C"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"103786","DOI":"10.1016\/j.mechmachtheory.2020.103786","article-title":"Vibration response and fault characteristics analysis of gear based on time-varying mesh stiffness","volume":"148","author":"Meng","year":"2020","journal-title":"Mech. Mach. Teory."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"462528","DOI":"10.1155\/2015\/462528","article-title":"Integrating the symmetry image and improved sparse representation for railway fastener classification and defect recognition","volume":"2015","author":"Liu","year":"2015","journal-title":"Math. Probl. Eng."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"61604","DOI":"10.1109\/ACCESS.2019.2961686","article-title":"Real-time inspection system for ballast railway fasteners based on point cloud deep learning","volume":"8","author":"Cui","year":"2020","journal-title":"IEEE Access"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Sun, X., Gu, J., Huang, R., Zou, R., and Palomares, B.G. (2019). Surface defects recognition of wheel hub based on improved Faster R-CNN. Electronics, 8.","DOI":"10.3390\/electronics8050481"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"125109","DOI":"10.1109\/ACCESS.2020.3007603","article-title":"SCueU-Net: Efficient damage detection method for railway rail","volume":"8","author":"Lu","year":"2020","journal-title":"IEEE Access"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Yuan, H., Chen, H., Liu, S., Lin, J., and Luo, X. (2019, January 14\u201317). A deep convolutional neural network for detection of rail surface defect. Proceedings of the 2019 IEEE Vehicle Power and Propulsion Conference, Hanoi, Vietnam.","DOI":"10.1109\/VPPC46532.2019.8952236"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Faghih-Roohi, S., Hajizadeh, S., N\u00fa\u00f1ez, A., Babuska, R., and Schutter, B.D. (2016, January 24\u201329). Deep convolutional neural networks for detection of rail surface defects. Proceedings of the 2016 International Joint Conference on Neural Networks, Vancouver, BC, Canada.","DOI":"10.1109\/IJCNN.2016.7727522"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Yanan, S., Hui, Z., Li, L., and Hang, Z. (December, January 30). Rail Surface Defect Detection Method Based on YOLOv3 deep learning networks. Proceedings of the 2018 Chinese Automation Congress, Xi\u2019an, China.","DOI":"10.1109\/CAC.2018.8623082"},{"key":"ref_24","unstructured":"Bochkovskiy, A., Wang, C.Y., and Liao, H.Y.M. (2020). YOLOv4: Optimal speed and accuracy of object detection. arXiv."},{"key":"ref_25","unstructured":"Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Ren, S., He, K., Girshick, R., and Sun, J. (2016). Faster R-CNN: Towards real-time object detection with region proposal networks. arXiv.","DOI":"10.1109\/TPAMI.2016.2577031"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23\u201328). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1007\/s11263-013-0620-5","article-title":"Selective search for object recognition","volume":"104","author":"Uijlings","year":"2013","journal-title":"Int. J. Comput. Vis."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Zitnick, C.L., and Doll\u00e1r, P. (2014, January 6\u201312). Edge boxes: Locating object proposals from edges. Proceedings of the European Conference on Computer Vision, Zurich, Switzerland.","DOI":"10.1007\/978-3-319-10602-1_26"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Chollet, F. (2017, January 21\u201326). Xception: Deep learning with depthwise separable convolutions. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.195"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/23\/11\/1437\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T07:23:24Z","timestamp":1760167404000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/23\/11\/1437"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,10,30]]},"references-count":30,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2021,11]]}},"alternative-id":["e23111437"],"URL":"https:\/\/doi.org\/10.3390\/e23111437","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,10,30]]}}}