Dynamics-Based Video Analytics
This research effort aims to substantially enhance our ability to exploit surveillance camera networks to predict and isolate threats from explosive devices in heavily crowded public spaces, and to guide complementary detection modalities, subsequent to a threat alert. At its core is a novel approach, stressing dynamic models as a key enabler for automatic, real time interpretation of what is currently an overwhelming profusion of video data streams. The project includes both theory development in an emerging new field, and an investigation of implementation issues.
A successful “who, doing what, where and why system,” can provide:
- faster throughput in airport security lines without compromising security,
- avoiding the of closure of airport terminals due to breach of security incidents (such as a person reaching the secure gates area through an exit, thus bypassing security),
- quick identification of recurrent thieves in public transportation terminals, and
- faster forensic analysis of security incidents.
All of these applications do not only have a tangible effect in ensuring public safety, but also have clear economic benefits such as reducing human resources needed at airport security checkpoints and reducing crime in bus terminals.
Our goal is to address the user needs for surveillance of large public spaces, such as airport terminals and bus stations. As part of this research, we are developing video analytics algorithms and implementing prototype systems, which are being tested using real-world data to show their feasibility.Year 4 Annual Report
Students Currently Involved in Project
- Sadjad Asghari-Esfeden
- Armand Comas
- Wenqian Liu
- Bengizu Ozbay
- Timothy Rupprecht
- Can Uner
- Dong Yin
- Yuexi Zhang
- Dan Luo