Intelligent Recognition Method for Mast Position of Overhead Contact Line

被引：6
|
作者：
Wang H. [1 ]
Zhou W. [1 ]
Zhang W. [1 ]
Li X. [1 ]
机构：
[1] Infrastructure Inspection Research Institute, China Academy of Railway Sciences Corporation Limited, Beijing
来源：
Zhongguo Tiedao Kexue/China Railway Science | 2019年 / 40卷 / 01期
关键词：
Catenary; Corner detection; Decision tree; Mast position; Random forest; Recognition; SPRINT algorithm; Stagger;
D O I：
10.3969/j.issn.1001-4632.2019.01.15
中图分类号：
学科分类号：
摘要：
The inspection data of catenary stagger within a certain distance is regarded as a binary image composed of kilometer post and stagger. At first, the stagger curve is transformed for noise reduction to lower the influence of the stagger difference between the adjacent points of contact wire on the selection of contour support domain during corner detection. Then the candidate mast position is initially detected by corner detection method based on sliding rectangle, utilizing the characteristic of the stagger curve that the stagger is symmetrical to the perpendicular line of the central line at most mast positions, ignoring the variations of the angle in the neighborhood of the measured point on the stagger curve, and considering only the corner points in the vertical direction. Finally, on the basis of extracting the feature attribute vectors of all candidate mast positions, and based on SPRINT decision tree algorithm, the random forest algorithm is adopted for the classification and intelligent recognition of the correct mast positions. Accordingly, the method is tested and verified. Results show that the proposed intelligent recognition method for mast position can identify the correct mast positions, and it has higher recognition accuracy under the premise of guaranteeing performance. © 2019, Editorial Department of China Railway Science. All right reserved.
引用
收藏
页码：111 / 116
页数：5
相关论文
共 11 条
  • [1] Zhou W., Sun Z., Ren S., Et al., Measurement Method for Geometric Parameters of Overhead Contact Line Based on Multi-View Stereovision, China Railway Science, 36, 5, pp. 104-109, (2015)
  • [2] Xu L., Research on Corner Detection Algorithm Based on Image Contour, (2009)
  • [3] Yang H., Hou C., Corner Detection Based on Distance, Angle and Template, Laser & Infrared, 35, 7, pp. 527-530, (2005)
  • [4] Verkeenko M.S., Development of an Algorithm for Fast Corner Points Detection, Journal of Computer and Systems Sciences International, 53, 3, pp. 392-401, (2014)
  • [5] Farzin M., Riku S., Robust Image Corner Detection through Curvature Scale Space, IEEE Transactions on Pattern Analysis and Machine Intelligence, 20, 12, pp. 1376-1381, (1998)
  • [6] Farzin M., Alan K.M., A Theory of Multiscale, Curvature-Based Shape Representation for Planar Curves, IEEE Transactions on Pattern Analysis and Machine Intelligence, 14, 8, pp. 789-805, (1992)
  • [7] Asif M., Sarfraz M., Corner Detection by Sliding Rectangles along Planar Curves, Computers & Graphics, 31, 3, pp. 440-448, (2007)
  • [8] Leo B., Random Forests, Machine Learning, 45, 1, pp. 5-32, (2001)
  • [9] Vladimir S., Andy L., Christopher T., Et al., Random Forest: a Classification and Regression Tool for Compound Classification and QSAR Modeling, Journal of Chemical Information and Computer Sciences, 43, 6, pp. 1947-1958, (2003)
  • [10] Antanas V., Adas G., Marija B., Mining Data with Random Forests: A Survey and Results of New Tests, Pattern Recognition, 44, 2, pp. 330-349, (2011)
12