期刊
出版社
ASSOC COMPUTING MACHINERY
DOI: 10.1145/3243394.3243703
关键词
Airport security; camera networks; video analytics
资金
- U.S. Department of Homeland Security [2013-ST-061-ED0001-04]
- National Science Foundation [IIS-1318145, ECCS-1404163]
Automatic algorithms for tracking and associating passengers and their divested objects at an airport security screening checkpoint would have great potential for improving checkpoint efficiency, including flow analysis, theft detection, line-of-sight maintenance, and risk-based screening. In this paper, we present algorithms for these tracking and association problems and demonstrate their effectiveness in a full-scale physical simulation of an airport security screening checkpoint. Our algorithms leverage both hand-crafted and deep-learning-based approaches for passenger and bin tracking, and are able to accurately track and associate objects through a ceiling-mounted multi-camera array. We validate our algorithm on ground-truthed datasets collected at the simulated checkpoint that reflect natural passenger behavior, achieving high rates of passenger/object/transfer event detection while maintaining low false alarm and mismatch rates.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据