Two-Dimensional CNN Based on Vertical Edge Detection and Density-Based Spatial Clustering for Φ-OTDR Vibration Recognition
A 2D-CNN utilizing vertical edge detection and density-based spatial clustering is proposed for Phase-sensitive OTDR vibration recognition. Field experiments demonstrate that our scheme can achieve 1m vibration positioning accuracy and 11.1% classification accuracy improvement. © 2024 The Author(s)
Authors
- Qian Hu (ORCID: https://orcid.org/0000-0002-8675-9637)
- Qiang Fu (ORCID: https://orcid.org/0000-0003-4100-2777)
- Xin Qin (ORCID: https://orcid.org/0000-0001-7457-0565)
- Yong Zhang (ORCID: https://orcid.org/0000-0001-6577-5235)
- Yan Zhang (ORCID: https://orcid.org/0000-0002-8561-5092)
- Yi Ding (ORCID: https://orcid.org/0000-0002-0046-3051)
- Jixiong Huang
- Xiankun Zhu
- Xia Gao
Institutions
- China Telecom (China) (CN)
- China Telecom (CN)
Publication Details
- Published
- 2024-08-04
- DOI
- https://doi.org/10.1109/cleo-pr60912.2024.10676483
- Citations
- 1
- Primary Topic
- Advanced Fiber Optic Sensors
- Type
- article
- Field-Weighted Citation Impact
- 0.21