The content of this article comes from the "Surveying and Mapping Bulletin" No. 3 of 2024, drawing review number: GS Jing (2024) No. 0499
Point cloud data is applied to the geometry detection of key nodes of turnouts
Wang Dongyan1,2, Yu Cai3, Shen Yan1, Zhang Zhenjian1,2, Li Yafeng1,2
1. Institute of Computing Technology, China Academy of Railway Sciences, Beijing 100081, China; 2. Beijing Jingwei Information Technology Co., Ltd., Beijing 100081, China; 3. Yinchuan Public Works Section, China Railway Lanzhou Bureau Group Co., Ltd., Yinchuan 750000, Ningxia, China
Funds: Research and Development Fund of China Academy of Railway Sciences Group Co., Ltd. (2022YJ288)
Keywords: 3D point cloud data, turnout, service status, convolutional neural network, laser scanning
Citation format: Wang Dongyan, Yu Cai, Shen Yan, et al. Application of Point Cloud Data in Geometric Detection of Key Nodes of Turnouts[J]. Bulletin of Surveying and Mapping, 2024(3): 107-112.doi: 10.13474/j.cnki.11-2246.2024.0318
Abstract:The turnout service state detection process is complex, and the traditional method needs to be combined with the rail inspection car, the gauge ruler, the branch gauge ruler and the reduced value measuring instrument for measurement, and the equipment types are complex, the measurement process is many, and the skylight time requirements are high. In order to improve the detection efficiency, this paper proposes a point cloud-based measurement method for the key nodes of single-open turnouts. The method uses CAD element information and graph convolutional neural network to realize the accuracy, automatic identification, segmentation and extraction of 3D point cloud data of turnout structure, with an accuracy of 99.68%. At the same time, combined with the geometric prior information of the turnout structure, the key geometric parameters such as the track gauge, the guide curve spacing, and the sharp rail reduction value of the turnout structure were quickly and accurately extracted. The example verification shows that the error between the measurement results and the true value of the key geometric shape and position detection method of turnouts based on point cloud proposed in this paper is sub-millimeter, which meets the requirements of actual engineering detection, saves a variety of detection equipment, saves a lot of skylight time, and has high practicability, which is the development trend of turnout measurement in the future.
About author:WANG Dongyan (1992—), female, master's degree, engineer, mainly engaged in railway public works informatization. E-mail:[email protected]
First instance: Yang Ruifang review: Song Qifan
Final Judge: Jin Jun
information