With the rapid increase of railway mileage, especially in complex geomorphic environments, it is necessary to use unmanned aerial vehicle (UAV) to automated monitor railway environmental security. This paper presents a railway intrusion detection method based on fast feature extraction and matching. Firstly, a low-degree polynomial detector (ALP) feature-based template database indexed by geolocation is established. Secondly, ALP descriptors are extracted in the region of interest (ROI) from the detection system for the sequences of railway images by the onboard camera of the UAV captured. Finally, ALP descriptors are matched between those from the real-time images and those from the feature database in the same geolocation and check whether there is an intrusion according to the matching ratio. The experiment results show that the proposed method can detect the invader effectively due to the obvious decrease of matching ratio and the successful detection ratio can up to 96%.
CITATION STYLE
Guan, L., Li, X., & Jia, L. (2020). An Intrusion Detection Method for Railway Based on Fast Feature Extraction and Matching of UAV Camera. In Lecture Notes in Electrical Engineering (Vol. 639, pp. 345–353). Springer. https://doi.org/10.1007/978-981-15-2866-8_33
Mendeley helps you to discover research relevant for your work.