Rail Joint Gap Measurement Method using Train Frontal Images Captured by a Handy Video Camcorder

被引：0
作者：
Goda W. [1 ]
Itoi K. [1 ]
Nagamine N. [1 ]
Tsubokawa Y. [1 ]
机构：
[1] Railway Technical Research Institute, 2-8-38, Hikari-cho, Kokubunji, Tokyo
来源：
IEEJ Transactions on Industry Applications | 2023年 / 143卷 / 01期
关键词：
deep learning; image processing; joint gap measurement; track maintenance;
D O I：
10.1541/ieejias.143.46
中图分类号：
学科分类号：
摘要：
Joint gap measurement is an important inspection to prevent buckling of railways. However, this measurement is labor intensive and time consuming because the tracks are too long for manual inspections. Therefore, we have developed a rail joint gap measurement method using train frontal images recorded by a handy video camcorder. In this paper, we describe details pertaining to the proposed method and discuss the measurement results obtained based on an experiment involving a commercial train. ©c 2023 The Institute of Electrical Engineers of Japan.
引用
收藏
页码：46 / 55
页数：9
相关论文
共 17 条
  • [1] 62, 12, pp. 35-37, (2008)
  • [2] pp. 403-404, (2008)
  • [3] Inagaki A., Ogino G., Shinoda N., Oomori Y., Shiono K., Improvement of the joint gaps measuring system, J-RAIL2001, pp. 655-658, (2001)
  • [4] Itoi K., Nagamine N., Goda W., Tsubokawa Y., Ohba H., Kato S., Application of Joint Gaps Measurement Method to Forward View Image from Train Cab, (2020)
  • [5] Fox-Ivey R., Nguyen T.V., Laurent J., Laser Triangulation for Track Change and Defect Detection, pp. 1-24, (2020)
  • [6] Guo F., Qian Y., Wu Y., Leng Z., Yu H., Automatic railroad track components inspection using real-time instance segmentation, Computer-Aided Civil and Infrastructure Engineering, 36, 3, pp. 362-377, (2021)
  • [7] Guo F., Qian Y., Shi Y., Real-time railroad track components inspection based on the improved YOLOv4 framework, Automation in Construction, 125, (2021)
  • [8] Wang T., Yang F., Tsui K.L., Real-Time Detection of Railway Track Component via One-Stage Deep Learning Networks, Sensors, 20, 15, (2020)
  • [9] (2015)
  • [10] (2011)
12