Engineers at the 麻豆原创 are devising a way to use video to detect abnormal behavior in large crowds in an effort to help keep people safe from terrorist attacks.
Professor Mubarak Shah heads the team that is using video streams, algorithms and a unique analysis to figure out how crowds behave and what might be suspicious behavior.
Video cameras monitoring people in public areas are now commonplace, and in large settings such as the Super Bowl, political rallies or royal weddings, the stakes are even higher because there are more people, which means more potential casualties.
Several research efforts are under way to develop systems that cue security personnel to individuals or events of interest in crowded scenes. Most of that now is handled by people who can easily miss something because it is impossible for them to monitor every change in the midst of a large crowd.
Shah is trying to automate most of the work with computer algorithms, and he鈥檚 modeling his work on how liquids behave in motion.
鈥淲hile human psychology and individual quirks complicate the analysis, in essence people in high-density crowds appear to move with the flow of the crowd, like particles in a liquid flow,鈥� Shah said.
So his team is building a program that analyzes behavior in a crowd much like analyzing particles and how they act in a fluid state. But there is one variation.
鈥淲e are saying one difference between crowds and fluid is that people have some destination they are going towards. That鈥檚 why crowds can be called 鈥榯hinking fluid,鈥欌�� Shah said.
The team鈥檚 work is promising and was the cover story last month in Communications of ACM, a computer science periodical.
鈥淲e鈥檙e still years away from perfecting this,鈥� Shah said. 鈥淏ut by using the basics of hydrodynamics we are developing a really good tool here.鈥�
Most work in automatic-video surveillance is done with crowds of no more than 20 people, Shah said, but if perfected could be helpful in crowds of hundreds or thousands. Shah鈥檚 research team used public video from large events such as the New York Marathon, a political rally in Los Angeles and pilgrims circling the Kaaba shrine in Mecca.
Computational and applied mathematics are critical for visual analysis of crowds, Shah added.
Pixel information from the screen is translated into particle trajectories used to understand crowd flow and then examined in various ways to recognize crowd behaviors, track individuals and detect unusual behavior.
Shah is an expert in the field of visual analysis. When he joined 麻豆原创 in 1986 he founded the Computer Vision Lab, which is a leader in developing technology for use in crowd surveillance, visual tracking, Unmanned Aerial Vehicle video analysis and analysis of crowded scenes.
Shah has a Ph.D. from Wayne State University and is a fellow of IEEE, AAAS, IAPR and SPIE, the largest professional organizations in his field. Several professional organizations have designated Shah as a distinguished speaker. He also has published hundreds of articles.聽 According to Academic Search, Shah is the sixth most frequently cited author in the world in Computer Vision in the past five years. He鈥檚 been honored with several awards, including Pegasus Professor, the highest award given by 麻豆原创 to a faculty member that has made a significant impact on the university. He also is a member of 麻豆原创鈥檚 Millionaires club. He teaches several courses, is director of the Computer Vision Lab and is the Agere Chair Professor of Computer Science.
Shah鈥檚 team includes Brian Moore, an assistant professor of mathematics at 麻豆原创; Saad Ali, a 麻豆原创 Computer Vision Lab alumnus and a computer scientist at SRI International in Princeton, N.J., and Ramin Mehran, who completed his Ph.D. 聽last month at 麻豆原创 and is joining Microsoft in January.