Back to Search Start Over

Research on Intelligent Monitoring Technology of Railway Operation and Maintenance Environment Based on UAV Platform.

Authors :
FENG Yao
Source :
Railway Investigation & Surveying; 2024, Vol. 50 Issue 5, p50-72, 8p
Publication Year :
2024

Abstract

Considering the disadvantages of traditional railway operation and maintenance manual inspection, including low efficiency, high cost, and detection blind spots, to solve the problems faced by drones in data acquisition, data processing and application, this paper innovatively proposed a route planning algorithm with a safe buffer zone. The algorithm adopted a partition operation strategy, and used two methods of orthography and side-view shooting to effectively improve the operation efficiency and the security of data acquisition. In addition, based on the data of UAV inspection aerial photography project, this paper formed a typical disease sample library in railway operation and maintenance environment through relevant processing, and used YOLOv7 target detection algorithm to realize the rapid automatic detection of three typical diseases of color steel tile, dustproof net and plastic film. Combined with human-computer interaction, the intelligent extraction and rapid filing of external environmental hazards were realized, and a set of intelligent monitoring technology scheme of railway operation and maintenance environment based on UAV platform was formed. The results show that, the effective combination of UAV technology, target detection technology and computer technology can realize rapid automatic hidden danger identification, which can improve the efficiency of railway operation and maintenance inspection by 150%, and provide important technical support for railway safety management. [ABSTRACT FROM AUTHOR]

Details

Language :
Chinese
ISSN :
16727479
Volume :
50
Issue :
5
Database :
Complementary Index
Journal :
Railway Investigation & Surveying
Publication Type :
Academic Journal
Accession number :
180331083
Full Text :
https://doi.org/10.19630/j.cnki.tdkc.202403260004