14 December 2023 in IEEE Transactions on Aerospace and Electronic Systems
Intelligent visual surveillance (IVS) is being gradually introduced into the field of airport surface surveillance. The first task of IVS is to detect and recognize objects moving on the airport surface. Specialty vehicles play a considerable role in airport ground handling processes and are considerably monitored targets. Because specialty vehicles have diverse appearances and irregular shapes, pixel-level detection would enable them to be targeted more accurately. Specialty vehicles on the surface undertake different jobs in airport ground handling processes, and therefore sub-category classification would more precisely determine the function of these specialty vehicles. Moreover, pixel-level detection and sub-category classification are very useful for detecting key milestone nodes of airport ground handling processes. Thus, in this paper, a two-stage framework for specialty vehicle pixel-level detection and sub-category classification for IVS of the airport surface is exploited, which seamlessly integrates state-of-the-art algorithms and techniques, and consists of two segmentation stages (coarse mask generation, and refined mask generation). Furthermore, to evaluate related methods, a dataset of airport surface specialty vehicles is established, which contains four types of representative specialty vehicles and corresponding accurate mask labels. All samples in the dataset were captured from surveillance videos of civil airports. Experimental results on the dataset clearly demonstrate that the proposed framework performed favorably compared with the classic instance segmentation methods, and achieved pixel-level detection and sub-category classification of specialty vehicles for airport surface surveillance.
11 February 2019 in Journal of Advanced Transportation
Quality of travel service for road transport relies heavily on richness of transport operation data. Currently, most types of data including coach operation data are collected by manual investigation which is time-consuming and labor-intensive, and this significantly hinders the realization of intelligent traffic information service. In view of the above problems, this paper is aimed at introducing a method of automatically extracting coach operation information using historical GPS trajectory data of massive coaches. The method first analyzes trajectory characteristics of coaches within stations and identifies the highly dense point clusters as coach stations using the DBSCAN clustering algorithm. Then the schedule information is obtained by conducting error adjustment on the actual arrival and departure time series of multiple shifts, and the name of coach station is queried from point of interest (POI) and geographical name database provided by online map. Finally, the regular driving route of coaches is extracted by an incremental trajectory merging method. The proposed method is applied in handling historical trajectory data in the Beijing-Tianjin-Hebei region in China, and experimental results show that the extraction accuracy is 84% and verify its effectiveness and feasibility. The proposed method makes use of data mining techniques to extract coach operation information from big trajectory data and saves a lot of labor work, time, and economic cost required by on-site investigation.