Study on rail gauge detection systems based on image sensors

MIN Yongzhi; WANG Hongxia; KANG Fei; DANG Jianwu

doi:10.7510/jgjs.issn.1001-3806.2015.03.014

Sensors are widely used in railway safety maintenance. In order to achieve real-time online gauge measurement, an on-board machine vision system of gauge detection was established with the combination of laser technique and CCD image sensors. Firstly, the working principle and structure of the system were introduced. A fast extraction method of rail contour centerline was proposed based on pre-noising and distance transform (PNDT) algorithm. Then, the image was enhanced with the strong contrast stretching method and exponential transform method. Fast pre-noising processing was conducted with the combination of Gaussian smoothing and dynamic region of interest (ROI). Finally, precise threshold segmentation processing was carried out for the images. The rail contour centerline was gotten and the gauge measurement point was located with the distance transform method. Experimental results show that the detection accuracy of the system is -1mm~+1mm and image frame processing speed is 14.35ms. The system can meet the requirement of the robustness, detection speed and accuracy in real-time online gauge detection.

Study on rail gauge detection systems based on image sensors

Corresponding author: DANG Jianwu, dangjw@mail.lzjtu.cn

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

Received Date: 2014-04-21

Accepted Date: 2014-05-20

Available Online: 2015-05-25

Abstract: Sensors are widely used in railway safety maintenance. In order to achieve real-time online gauge measurement, an on-board machine vision system of gauge detection was established with the combination of laser technique and CCD image sensors. Firstly, the working principle and structure of the system were introduced. A fast extraction method of rail contour centerline was proposed based on pre-noising and distance transform (PNDT) algorithm. Then, the image was enhanced with the strong contrast stretching method and exponential transform method. Fast pre-noising processing was conducted with the combination of Gaussian smoothing and dynamic region of interest (ROI). Finally, precise threshold segmentation processing was carried out for the images. The rail contour centerline was gotten and the gauge measurement point was located with the distance transform method. Experimental results show that the detection accuracy of the system is -1mm~+1mm and image frame processing speed is 14.35ms. The system can meet the requirement of the robustness, detection speed and accuracy in real-time online gauge detection.

HTML

Reference (17)

[1]	SHOLL H, AMMAR R, GREENSHIELDS I, et al. Application of computing analysis to real-time railroad track inspection[C]//Automation Congress. New York, USA:IEEE,2006:1-6.
[2]	HUANG J X. The application study of image processing in track detection[D]. Beijing: Beijing Jiaotong University, 2007:33-83(in Chinese).
[3]	LIU T, REN S W, XU G Y, et al. Upgrading of gauge-alignment system of type GJ-4 track inspection car[J]. China Railway Science, 2006, 32(6):137-140(in Chinese).
[4]	RESENDIZ E, HART J M, AHUJA N. Automated visual inspection of railroad track[J].Intelligent Transportation Systems, 2013, 14(2):751-760.
[5]	RIZZO P, CAMMARATA M, BARTOLI I. Ultrasonic guided waves-based monitoring of rail head laboratory and field tests[J]. Advances in Civil Engineering, 2010, 6(10):1-13.
[6]	HU K, ZHOU F Q, ZHANG G J. An rapid sub-pixel accuracy extraction method for structured light stripe center[J]. Chinese Journal of Science Instrument, 2006, 27(10):1326-1329(in Chinese).
[7]	LEI H J, LI D H, WANG J Y, et al. An rapid detection method structured light stripe center[J]. Huazhong University of Science and Technology (Natural Science Edition), 2003, 31(1):74-76(in Chinese).
[8]	LI Z W, WANG C J, SHI Y S. An extraction algorithm for combination of gradient sharpening and optical center of gravity method[J]. Chinese Journal of Image and Graphics, 2008, 13(1): 64-68(in Chinese).
[9]	LI M, FENG H J, XU Z H, et al. A corrected image contour extraction method using intensity information for structured light[J]. Optical Engineering, 2005, 32(2):30-32(in Chinese).
[10]	ZHAN L, YU L, XIAO J, et al. Calibration method research of laser camera sensor in track detection[J]. Journal of Mechanical Engineering, 2013, 49(16):39-47(in Chinese).
[11]	MI C Z, XIE Z J, CHEN T, et al. The key technology of image enhancement and edge extraction of heavy rail[J]. Optics and Precision Engineering, 2012, 20(7):1645-1652(in Chinese).
[12]	OUYANG Ch S, YUAN J, TIAN J W, et al. An enhancement method for X-ray image using rough sets and human visual system [J]. Journal of Xi'an Jiaotong University, 2009, 43(6): 48-51(in Chinese).
[13]	SONKA M, HLAVAC V, BOYLE R. Image processing analysis and machine vision[M].3rd ed. Beijing: Tsinghua University Press, 2007:1-613(in Chinese).
[14]	ZHAO B H, WANG B X, ZHANG J, et al. Center extraction method for rough metal surface light strip[J]. Optics and Precision Engineering, 2011, 19(9):2138-2145(in Chinese).
[15]	ZHAO J, ZHAO J, ZHANG L. Image processing and feature extraction for structured light images of welded seam [J]. Journal of Xi'an Jiaotong University, 2013, 47(1):114-119(in Chinese).
[16]	WANG M J, YANG L, WANG X, et al.The class euclidean distance transformation class features extraction with a simplified pulse coupled neural network traffic sign image[J]. Optics and Precision Engineering, 2012, 20(12):2751-2758(in Chinese).
[17]	SAPIRO G. Geometric partial differential equations and image analysis[M].3rd ed. London,UK: Cambridge University Press, 2006:1-532.

Study on rail gauge detection systems based on image sensors

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Proportional views