Intelligent Upgrading and Application of Bridge Video Surveillance System Based on Computer Vision
|
|
|||||||||||
Bibliographic Details
| Author(s): |
Yuan Chen
(CCCC Second Harbor Engineering Company LTD, Wuhan, China)
Xiankai Xue (CCCC Second Harbor Engineering Company LTD, Wuhan, China) Yongwei Wang (CCCC Second Harbor Engineering Company LTD, Wuhan, China; Key Laboratory of Large-span Bridge Construction Technology, Wuhan, China) Hao Zhu (CCCC Second Harbor Engineering Company LTD, Wuhan, China; Key Laboratory of Large-span Bridge Construction Technology, Wuhan, China) |
||||
|---|---|---|---|---|---|
| Medium: | conference paper | ||||
| Language(s): | English | ||||
| Conference: | IABSE Congress: Bridges and Structures: Connection, Integration and Harmonisation, Nanjing, People's Republic of China, 21-23 September 2022 | ||||
| Published in: | IABSE Congress Nanjing 2022 | ||||
|
|||||
| Page(s): | 1147-1153 | ||||
| Total no. of pages: | 7 | ||||
| DOI: | 10.2749/nanjing.2022.1147 | ||||
| Abstract: |
The rapid development of computer vision provides a foundation for the intelligent upgrading of bridge video surveillance systems. In this paper, two intelligent upgrading methods were developed and deployed. The first method uses edge computing equipment as the core, to quickly identify and locate vehicles across the large-span bridge by YOLOv5, which was trained by synthesized vehicle dataset, and then a large-span bridge vehicle digital twin system was built and deployed in Baijusi Yangtze River Bridge, which is suitable for scenarios with high real-time requirements. The another one is based on cloud computing, relying on ShuffleNetV2 to build a waterlogging recognition model and early warning system, which is suitable for scenarios with low real-time requirements. The results show that the constructed intelligent system upgrades the traditional passive access system to an early warning system with active recognition, which improves the intelligence of the system and meets the needs of engineering applications. |
||||
| Keywords: |
early warning deep learning video surveillance intelligent upgrading
|
||||
| Copyright: | © 2022 International Association for Bridge and Structural Engineering (IABSE) | ||||
| License: | This creative work is copyrighted material and may not be used without explicit approval by the author and/or copyright owner. |
||||