Nabil J. Sarhan, a professor of electrical and computer engineering at Detroit’s Wayne State University, today announced he had a patent approved for an automated video surveillance system. The system is monitored entirely by computers.
Most large-scale security systems include multiple, changing displays monitored by people, so crimes can be missed if people are distracted or the display is momentarily unavailable. Footage often is used as evidence after a crime. Sarhan’s system works live — it constantly scans for potential threats while allocating resources to optimize overall threat detection.
“There are so many considerations that we take and then solve mathematically to get the video recordings in the best way possible,” says Sarhan. “It’s unique.”
The system uses facial and license plate recognition on high-quality video that computers analyze and allows for certain areas observed by the cameras to be placed at higher levels of importance. It offers an increase in system coverage with reduced cost due to lower bandwidth and energy needs.
“We’re applying a lot of similar principles to internet video streaming as we are to this security system, such as the idea of being smarter about how we use our resources,” says Sarhan. “To be using this knowledge for security systems, however, feels like a more worthwhile pursuit.”
Sarhan’s project stemmed from his research in designing multimedia systems. He specializes in multimedia, computer systems and networks, and computer architecture.
The system is being deployed in the College of Engineering. Sarhan and his team received a $290,000 National Science Foundation grant in the early stages of the nearly eight-year project.