A Low-Complexity Accurate Ranging Algorithm for a Switch Machine Working Component Based on the Mask RCNN
Source abstract
Abstract
According to the intelligent development needs of railway operation and maintenance, turnout maintenance also needs an efficient and intelligent means of detection. It is the main method used to measure the access depth of static contact manually. In order to change the disadvantages of the low efficiency and strong subjectivity of traditional schemes, a low-complexity accurate ranging algorithm of the Mask RCNN is proposed to measure the on–off working parts. Firstly, the Mask RCNN and an interactive iterative method are used to segment the region of interest accurately twice. Secondly, the graph distortion is corrected according to the vertex mapping principle. Finally, the accurate actual distance is calculated through fitting the linear distance transformation equation. Through the secondary segmentation and correction algorithm, the accurate calculation of a small target is completed. The experimental results show that the algorithm can accurately measure the distance of different working parts; the average processing time is 0.8 s/amplitude and the measurement error is ±1 mm.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Surface Roughness and Optical Measurements · Image and Object Detection Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Urban Railway Transportation, Shanghai University of Engineering Science, Shanghai 201620, China
References (14)
Show source-provided references
1. Defect Detection of Pandrol Track Fastener Based on Local Depth Feature Fusion Network10.1155/2021/6687146 ↗
2. Ultrasonic Distance Measurement Method With Crosstalk Rejection at High Measurement Rate10.1109/TIM.2018.2863999 ↗
3. High-Accuracy Ultrasonic Rangefinders via pMUTs Arrays Using Multi-Frequency Continuous Waves10.1109/JMEMS.2019.2912869 ↗
4. Fringe Disappearance in Self-Mixing Interferometry Laser Sensors: Model and Application to the Absolute Distance Measurement Scheme10.1109/JSEN.2019.2901125 ↗
5. A review and discussion on deep learning
6. Algorithm of Speed-up Turnout Fault Diagnosis Based on BP Neural Network
7. Deep neural networks: A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data10.1016/j.ymssp.2015.10.025 ↗
8. Gap Detection of Switch Machines in Complex Environment Based on Object Detection and Image Processing10.1061/JTEPBS.0000406 ↗
9. Automatic Detection Algorithm of Switch Machine Gap Based on Canny Operator
10. A Railway Turnout Detection Method Based on Machine Vision
11. He, K., Gkioxari, G., Dollar, P., and Girshick, R. (2017, January 22–29). Mask RCNN. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy.10.1109/ICCV.2017.322 ↗
12. Positioning of the Moving and Static Contacts of the Switch Machine Based on Double-Layer Mask RCNN10.1155/2021/8655499 ↗
13. Zhang, C., and Xu, Q. (2019, January 3–5). An Improved GMM based Video Foreground Separation. Proceedings of the 2019 Chinese Control And Decision Conference (CCDC), Nanchang, China.10.1109/CCDC.2019.8832566 ↗
14. Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C.L. (2014, January 6–12). Microsoft COCO: Common Objects in Context. Proceedings of the Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland.10.1007/978-3-319-10602-1_48 ↗