Research on Tracking Foreign Objects in Railway Tracks Based on Hidden Markov Kalman Filter

Hou Tao; Zhao Yanzhang; Niu Hongxia; Chen Mingxi; Wang Shan

doi:10.6180/jase.202210_25(5).0020

Research on Tracking Foreign Objects in Railway Tracks Based on Hidden Markov Kalman Filter

Hou Tao¹, Zhao Yanzhang This email address is being protected from spambots. You need JavaScript enabled to view it.¹, Niu Hongxia¹, Chen Mingxi¹, and Wang Shan¹

¹School of Automation and Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China

Received: September 15, 2021
Accepted: December 6, 2021
Publication Date: January 19, 2022

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202210_25(5).0020

ABSTRACT

The intrusion of foreign object on the railway tracks directly affects the safe operation of the trains. Due to the complex railway tracks environment, the existing research on foreign object detection based on image processing has problems such as weak anti-noise, poor real-time performance, and low accuracy. A real-time detection and tracking method for foreign bodies invading railway tracks based on Hidden Markov Model (HMM) Kalman Filter is put forward in this paper. Firstly, the Gaussian Mixture Model is used to extract the feature vector of the object in multiple images and generate a feature sequence. Secondly, the feature sequence of the detected object is processed by the Hidden Markov Model, and the movement railway tracks of the foreign object is predicted. Finally, the prediction result is compared with the actual results. The Kalman filter is updated according to the comparison results, and the foreign objects invading the railway tracks are finally detected and tracked. The simulation results show that the method can detect accurately and quickly and track foreign objects invading the railway tracks. Compared with the existing foreign object detection results achieved by the application of neural networks and Gaussian Mixture Models, the processing and results of the algorithm have strong anti-noise performance and real-time performance. It has high definition and an accuracy rate of 98.73%, which can further ensure the safety of train operation.

Keywords: Foreign object intrusion, Gaussian Mixture Model, Kalman Filter, Hidden Markov Model, Hungary algorithm

REFERENCES

[1] W. Yao, Y. Zujun, Z. Liqiang, and G. Baoqing, (2019) “Incursion Detection Method Based on Higher-order Fullyconnected Conditional Random Fields" Journal of the China Railway Society: DOI: 10.3969/j.issn.1001-8360.2019.05.010.
[2] H. Tao, W. Haiping, and N. Hongxia, (2020) “Realtime detection of rail dynamic foreign object intrusion based on improved MOG-LRMF" Transportation System Engineering and Information: 20(2):91–100. DOI:10.16097/j.cnki.1009-6744.2020.02.014.
[3] N. Zheng, N. Hongxia, and Z. Zhaoxin, (2021) “Railway foreign body intrusion detection method based on improved mixed Gaussian model" Sensors and Microsystems: 40(05): 146–149.
[4] X. Xinlong. Research on railway foreign object intrusion detection method based on target enhancement fusion. Beijing, China, 2020.
[5] C. Tastimur, M. Karakose, and E. Akin, (2017) “Image Processing Based Level Crossing Detection and Foreign Objects Recognition Approach in Railways" International Journal of Applied Mathematics, Electronics and Computers Special Issue-1:
[6] Y. X. Meng Yun. A Survey of Object Tracking Algorithms. Acta Automatica Sinica. 2019. DOI: 10.16383/j.aas.c180277.
[7] S. A. Memon, T. L. Song, K. H. Memon, I. Ullah, and U. Khan, (2020) “Modified smoothing data association for target tracking in clutter." Expert Systems With Applications (141): DOI: 10.1016/j.eswa.2019.112969.
[8] H. Suixin, C. Yanchang, Y. Jun, and Z. Zhenyuan, (2020) “Multi-Target Tracking Algorithm for Highway Vehicles Based on Multi- Feature Fusion" Automobile Technology (09): 8–13.
[9] C.Wei, Y.Wending, F. Qiang, L. Hai, and X. Qiqing, (2021) “Meanshift tracking algorithm based on adaptive kalman filter" Manufacturing Automation 43(06): 16–20.
[10] W. Sunyong, N. Qiaojiao, C. Ruhua, S. Xiyan, and P. Fubiao, (2019) “Small targets detection and tracking algorithm using box particle probability hypothesis density fifilter" Control and Decision 34(07): 1417–1424.
[11] Z. Yi, W. Mingzhou, and H. Youfeng, (2011) “Improved central differential Kalman filter for underwa ter passive target tracking" Ship Science and Technology 43(01): 154–160.
[12] W. Quandong, Y. Yue, L. Yiping,W. Xiaobin, and L. Sisi, (2019) “Review on railway intrusion detection methods" Journal of Railway Science and Engineering 16(12): 3152–3159.
[13] Y. Yuhan, H. Bin, C. Zheng, and M. Ying, (2020) “Algorithm for detecting moving targets at sea based on Gaussian mixture model" Application Research of Computers, 37(S1): 310–313.
[14] L. Huacai, H. Huazhan, H. Yiqing, and G. Wengen, (2019) “Improved canny edge operator and gaussian mixture model for moving target detection" Journal of Electronic Measurement and Instrument 33(10): 142–147.
[15] Z. Xiaorong, J. Huimin, andW. Yuning, (2019) “Power dependent structure modeling of photovoltaic power plants based on the combination of Gaussian mixture model and Copula function" Acta Solar Energy 40(7): 1912–1919.
[16] C. Jun, Y. Jiajia, and M. Jun, (2019) “Local Gaussian Distribution Fitting Energy Model with Fractional Differential" Pattern Recognition and Artificial Intelligence 32(5): 409–419. DOI: 10.16451/j.cnki.issn1003-6059.201905003.
[17] C. Luojia, F. Xinxi, and W. Quan, (2019) “Gaussian Particle Filter for Extended Target Tracking Based on Gaussian Process Regression" Journal of Projectiles, Rockets, Rockets and Guidance 39(2): 115–119.