{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T22:08:02Z","timestamp":1778278082059,"version":"3.51.4"},"reference-count":38,"publisher":"MDPI AG","issue":"21","license":[{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>The efficiency and the effectiveness of railway intrusion detection are crucial to the safety of railway transportation. Most current methods of railway intrusion detection or obstacle detection are inappropriate for large-scale applications due to their high cost or limited coverage. In this study, we present a fast and low-cost solution to intrusion detection of high-speed railways. As the solution to heavy computational burdens in the current convolutional-neural-network-based detection methods, the proposed method is mainly a novel neural network based on the SSD framework, which includes a feature extractor using an improved MobileNet and a lightweight and efficient feature fusion module. In addition, aiming to improve the detection accuracy of small objects, the feature map weights are introduced through convolution operation to fuse features at different scales. TensorRT is employed to optimize and deploy the proposed network in the low-cost embedded GPU platform, NVIDIA Jetson TX2, to enhance the efficiency. The experimental results show that the proposed methods achieved 89% mAP on the railway intrusion detection dataset, and the average processing time for a single frame was 38.6 ms on the Jetson TX2 module, which satisfies the need of real-time processing.<\/jats:p>","DOI":"10.3390\/s21217279","type":"journal-article","created":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T22:24:22Z","timestamp":1635805462000},"page":"7279","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":14,"title":["A Fast Intrusion Detection Method for High-Speed Railway Clearance Based on Low-Cost Embedded GPUs"],"prefix":"10.3390","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8288-9549","authenticated-orcid":false,"given":"Yao","family":"Wang","sequence":"first","affiliation":[{"name":"School of Mechanical, Electronic and Control Engineering, Beijing Jiaotong University, Beijing 100044, China"}]},{"given":"Peizhi","family":"Yu","sequence":"additional","affiliation":[{"name":"Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA"}]}],"member":"1968","published-online":{"date-parts":[[2021,11,1]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"702","DOI":"10.1109\/TITS.2010.2052101","article-title":"Efficient Multisensory Barrier for Obstacle Detection on Railways","volume":"11","author":"Hernandez","year":"2010","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Liu, Q., Qin, Y., Xie, Z., Yang, T., and An, G. (2017, January 20\u201322). Intrusion Detection for High-Speed Railway Perimeter Obstacle. Proceedings of the International Conference on Electrical and Information Technologies for Rail Transportation (EITRT), Changsha, China.","DOI":"10.1007\/978-981-10-7989-4_47"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"102811","DOI":"10.1109\/ACCESS.2020.2997946","article-title":"A Deep Learning Approach Towards Railway Safety Risk Assessment","volume":"8","author":"Alawad","year":"2020","journal-title":"IEEE Access"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Guo, B., Shi, J., Zhu, L., and Yu, Z. (2019). High-Speed Railway Clearance Intrusion Detection with Improved SSD Network. Appl. Sci., 9.","DOI":"10.3390\/app9152981"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"110","DOI":"10.2219\/rtriqr.46.110","article-title":"Level Crossings Obstacle Detection System Using Stereo Cameras","volume":"46","author":"Ohta","year":"2005","journal-title":"Q. Rep. Rtri"},{"key":"ref_6","unstructured":"Rodriguez, L.A., Uribe, J., and Bonilla, J.F.V. (2012, January 12\u201314). Obstacle detection over rails using hough transform. Proceedings of the 2012 XVII Symposium of Image, Signal Processing, and Artificial Vision, Medellin, Colombia."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Silar, Z., and Dobrovolny, M. (2013, January 2\u20134). The obstacle detection on the railway crossing based on optical flow and clustering. Proceedings of the 2013 36th International Conference on Telecommunications and Signal Processing, Rome, Italy.","DOI":"10.1109\/TSP.2013.6614039"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1177\/0361198118792751","article-title":"Video Analytics for Railroad Safety Research: An Artificial Intelligence Approach","volume":"2672","author":"Zaman","year":"2018","journal-title":"Transp. Res. Rec."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1177\/0361198119846468","article-title":"Artificial Intelligence-Aided Automated Detection of Railroad Trespassing","volume":"2673","author":"Zaman","year":"2019","journal-title":"Transp. Res. Rec."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ye, T., Wang, B., Song, P., and Li, J. (2018). Automatic Railway Traffic Object Detection System Using Feature Fusion Refine Neural Network under Shunting Mode. Sensors, 18.","DOI":"10.3390\/s18061916"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"145182","DOI":"10.1109\/ACCESS.2020.3015251","article-title":"Autonomous Railway Traffic Object Detection Using Feature-Enhanced Single-Shot Detector","volume":"8","author":"Ye","year":"2020","journal-title":"IEEE Access"},{"key":"ref_12","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 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Girshick, R. (2015, January 7\u201313). Fast r-cnn. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.","DOI":"10.1109\/ICCV.2015.169"},{"key":"ref_14","first-page":"91","article-title":"Faster r-cnn: Towards real-time object detection with region proposal networks","volume":"28","author":"Ren","year":"2015","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"040017","DOI":"10.1063\/1.5039091","article-title":"Railway obstacle detection algorithm using neural network","volume":"1967","author":"Yu","year":"2018","journal-title":"Aip Conf. Proc."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Redmon, J., and Farhadi, A. (2017, January 21\u201326). YOLO9000: Better, faster, stronger. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.690"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016, January 11\u201314). Ssd: Single shot multibox detector. Proceedings of the Computer Vision\u2014ECCV 2016, Amsterdam, The Netherlands.","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Xu, Y., Gao, C., Yuan, L., Tang, S., and Wei, G. (2019, January 27\u201330). Real-time Obstacle Detection Over Rails Using Deep Convolutional Neural Network. Proceedings of the 2019 IEEE Intelligent Transportation Systems Conference (ITSC), Auckland, New Zealand.","DOI":"10.1109\/ITSC.2019.8917091"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1375","DOI":"10.1109\/TITS.2020.2969993","article-title":"Railway traffic object detection using differential feature fusion convolution neural network","volume":"22","author":"Ye","year":"2020","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Hyde, P., Ulianov, C., Liu, J., Banic, M., Simonovic, M., and Ristic-Durrant, D. (2021). Use cases for obstacle detection and track intrusion detection systems in the context of new generation of railway traffic management systems. Proc. Inst. Mech. Eng. Part J. Rail Rapid Transit.","DOI":"10.1177\/09544097211041020"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"105791","DOI":"10.1016\/j.nanoen.2021.105791","article-title":"Solid ion channels gel battery driven by triboelectric effect and its integrated self-powered foreign matter intrusion detecting system","volume":"83","author":"Bi","year":"2021","journal-title":"Nano Energy"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Sun, Y., Xie, Z., Qin, Y., Chuan, L., and Wu, Z. (2020, January 6\u20139). Image Detection of Foreign Body Intrusion in Railway Perimeter Based on Dual Recognition Method. Proceedings of the European Workshop on Structural Health Monitoring, online.","DOI":"10.1007\/978-3-030-64908-1_60"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Kwon, H., Kim, Y., Yoon, H., and Choi, D. (2017). Optimal Cluster Expansion-Based Intrusion Tolerant System to Prevent Denial of Service Attacks. Appl. Sci., 7.","DOI":"10.3390\/app7111186"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1016\/j.inffus.2021.02.007","article-title":"On learning effective ensembles of deep neural networks for intrusion detection","volume":"72","author":"Folino","year":"2021","journal-title":"Inf. Fusion"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Mukojima, H., Deguchi, D., Kawanishi, Y., Ide, I., Murase, H., Ukai, M., Nagamine, N., and Nakasone, R. (2016, January 25\u201328). Moving camera background-subtraction for obstacle detection on railway tracks. Proceedings of the 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA.","DOI":"10.1109\/ICIP.2016.7533104"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Tastimur, C., Karakose, M., and Akin, E. (2017). Image processing based level crossing detection and foreign objects recognition approach in railways. Int. J. Appl. Math. Electron. Comput., 19\u201323.","DOI":"10.18100\/ijamec.2017SpecialIssue30465"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"245","DOI":"10.1007\/s00521-012-0846-0","article-title":"Efficient railway tracks detection and turnouts recognition method using HOG features","volume":"23","author":"Qi","year":"2013","journal-title":"Neural Comput. Appl."},{"key":"ref_28","unstructured":"Sun, Y., and Xie, Z. (2019, January 25\u201327). A Method for Pedestrian Intrusion Detection of Railway Perimeter Based on HOG and SVM. Proceedings of the 4th International Conference on Electrical and Information Technologies for Rail Transportation (EITRT) 2019, Qingdao, China."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1904","DOI":"10.1109\/TPAMI.2015.2389824","article-title":"Spatial pyramid pooling in deep convolutional networks for visual recognition","volume":"37","author":"He","year":"2015","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_30","unstructured":"Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., and LeCun, Y. (2013). Overfeat: Integrated recognition, localization and detection using convolutional networks. arXiv."},{"key":"ref_31","unstructured":"Fu, C.Y., Liu, W., Ranga, A., Tyagi, A., and Berg, A.C. (2017). Dssd: Deconvolutional single shot detector. arXiv."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017, January 22\u201329). Focal loss for dense object detection. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Zhang, S., Wen, L., Bian, X., Lei, Z., and Li, S.Z. (2018, January 18\u201323). Single-shot refinement neural network for object detection. Proceedings of the 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00442"},{"key":"ref_34","unstructured":"Jia, X., Sun, Y., and Xie, Z. (2019, January 25\u201327). Application of Target Detection Algorithms in Railway Intrusion. Proceedings of the 4th International Conference on Electrical and Information Technologies for Rail Transportation (EITRT) 2019, Qingdao, China."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"1071","DOI":"10.1007\/s00138-019-01038-4","article-title":"An embedded implementation of CNN-based hand detection and orientation estimation algorithm","volume":"30","author":"Yang","year":"2019","journal-title":"Mach. Vis. Appl."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Tijtgat, N., Van Ranst, W., Goedeme, T., Volckaert, B., and De Turck, F. (2017, January 22\u201329). Embedded real-time object detection for a UAV warning system. Proceedings of the 2017 IEEE International Conference on Computer Vision Workshops (ICCVW), Venice, Italy.","DOI":"10.1109\/ICCVW.2017.247"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Saypadith, S., and Aramvith, S. (2018, January 12\u201315). Real-time multiple face recognition using deep learning on embedded GPU system. Proceedings of the 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Honolulu, HI, USA.","DOI":"10.23919\/APSIPA.2018.8659751"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Yudin, D., and Slavioglo, D. (2018, January 10\u201314). Usage of fully convolutional network with clustering for traffic light detection. Proceedings of the 2018 7th Mediterranean Conference on Embedded Computing (MECO), Budva, Montenegro.","DOI":"10.1109\/MECO.2018.8406049"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/21\/7279\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T07:24:21Z","timestamp":1760167461000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/21\/7279"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,1]]},"references-count":38,"journal-issue":{"issue":"21","published-online":{"date-parts":[[2021,11]]}},"alternative-id":["s21217279"],"URL":"https:\/\/doi.org\/10.3390\/s21217279","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11,1]]}}}