Security cameras are everywhere today – at the mall, in the parking lot, even on a downtown street. If you’re on the right side of the law, you probably breathe a sigh of relief when you see them. After all, the images from those cameras are being beamed to a security room or police station somewhere. It’s nice to know that someone is watching out for our safety.
But are they really? Think about it; with so many cameras to watch, can any security guard – even a highly trained one – do an adequate job? Even hiring professional, dedicated guards doesn’t mean you’re getting top security, since research shows that even a highly motivated guard will lose his or her ability to concentrate on a monitor screen properly after about 20 minutes.
Ami Paz and Lizy Gedalyahu of Israeli high-tech company MATE Intelligent Video believe there’s good reason to be worried. The more cameras installed, the less attention the people who are supposed to be watching what they observe tend to pay; watching video monitors gets boring after awhile and guards can easily miss the subtle nuances that indicate a security breach.
“And when shifts change, there is a complete break in the observation of the flow of events, since a new person who was not aware of what went on before is now working in the observation booth – and as such, the new person won’t easily notice slight changes in the landscape that could indicate a security breach,” Gedalyahu explains to ISRAEL21c.
Same goes for cameras that transmit images back to a server for recording; someone has to actually sit and watch the footage in order to find out where the security breach was. “On average, observers watching events catch maybe one percent of what is actually going on in real time,” says Gedalyahu. “The best use of footage is usually for review of events that already happened – making most video detection technology reactive instead of proactive.”
MATE could be the solution. The company uses video analytics that can detect changes in the landscape, indicating if an intruder or unauthorized person has entered the area.
Instead of relying on an error-prone human to detect problems, MATE’s range of perimeter detection and video surveillance systems use sophisticated algorithms to constantly check what is happening in the field.
And when an unauthorized change is detected, the system alerts security personnel, pinpointing the problem and allowing them to efficiently take care of the situation before it gets out of hand.
MATE’s technology can be used for anything from detecting suspicious activities or behavior, to counting people and cars, tracking objects, or even detecting tailgating or piggybacking through access-controlled doors.
It reads the content of the images and analyzes them, then gives warning to specific events that get triggered. “It’s like a video sensor,” with an alarm that goes off when the sensor detects that something is awry, says Paz.
For example, says Paz, a MATE product might be deployed at the perimeter of an airport – an area for “authorized personnel only.” Given the security concerns at airports today, security people tend to take unauthorized entry into secure areas very seriously.
MATE products constantly compare images taken by the camera, frame by frame, and analyze the images for objects (like people) that shouldn’t be there.
“The MATE system knows what to look for, including surreptitious movement inside the authorized area. It can detect, for example, cars that try to ‘tailgate’ their way into an area, attempting to break in by shadowing a vehicle with a permit to enter,” Paz says.
“You remember those old cartoons where the bad guy sneaks around in a bush to get closer to his target? MATE will zero in on that moving bush right away, whereas a human observer might not be observant enough to notice that it moved,” he adds.
What about “foreign objects” that don’t belong, but don’t need to be stopped – like birds? “No problem,” says Paz. “MATE knows just what to look for and what to ignore. The system’s algorithms understand the video, and can detect wind, birds, cats, changes in lighting, and many other environmental factors. The rate of false alarms with the system is extremely low.”
What is extremely high, though, is the MATE video analytics accuracy rate. “There is no question that the MATE system is dozens of times more efficient and cost effective than the human observation system,” Paz asserts.
“Without MATE, you’d have to hire several people for round the clock shifts to observe sensitive installations,” says Paz. “With MATE, one person to manage the security system is probably enough. The other workers can be reassigned in a more productive manner.”