Intelligent video surveillance promises to boost security
effectiveness by automatically alerting personnel to take action when a
security violation occurs. In theory, this frees security guards from
the monotony of watching a large number of video displays. Use of such
"video analytics" leverages the inherent strengths of machines and
people: Automated sensors never tire, can cover large distances, and
"see" what the eye would miss, even in the absolute darkness. People can
then make smart response decisions. And for many indoor surveillance
applications, this is often the case.
When
it comes to outdoor surveillance, too often the reality is different.
Outdoors, the security officer who was supposed to be more efficient now
spends his time dealing with nuisance alarms that result when a gust of
wind or a change in lighting triggers the video detection system
inappropriately. As a result, security personnel come to distrust the
system, and may tune down the detection sensitivity or possibly even
turn off the alerts themselves.
Some facilities have hundreds of
nuisance alarms every week. One reason is that video intrusion detection
systems are often being used outdoors in applications for which they
weren't designed. The time security personnel spend addressing these
nuisance alarms can negate the claimed efficiency advantages of
intelligent video, while redirecting their efforts away from other areas
of security importance.
The key to eliminating nuisance alarms
outdoors is to use technology designed for outdoor applications rather
than mis-applying analytics intended for more controlled indoor
surroundings. Indoors, a camera only needs to see a limited field of
view in typically controlled surroundings. It's a mistake to apply the
same technologies to monitor critical infrastructure applications such
as transportation, energy, utilities or large campuses in the outdoors
where conditions are continually changing.
Consider the typical
outdoor surveillance scene: Cameras are mounted high on poles which
shake from even a slight wind or vibration. Clouds are moving across the
sky, creating shadows on the ground that appear to the camera as moving
objects. Trees and their leaves flap in the breeze, further creating
the appearance of moving objects. When you add in snow, rain, humidity
and dust, such a dynamic environment will wreak havoc for video
analytics not intended to operate under such conditions.
Using
intelligent video to secure large outdoor venues requires the use of
specific technologies like sufficient on-board camera processing power
to overcome lighting and weather issues and accurately detect and track
legitimate targets from extraneous surrounding motion and clutter. Such
systems can also employ geographic information system (GIS) coordinates
to determine a target's location, size and velocity, even over large
fields of view.
There are cost savings associated with such an
approach. That's because the same image processing that gives these
cameras their outdoor detection accuracy also gives them extended range.
This increased coverage translates to substantial project cost savings -
typically on the order of 50 percent. This is due to elimination of the
extra camera poles, construction, trenching, power and network
connectivity that would have been required.
Applying the right
technologies to video surveillance systems can transform the performance
and economics of outdoor security. Reductions in design, construction
and installation costs are also possible when truly intelligent
long-range cameras provide wide-area coverage with accurate data. The
resulting solution will be trusted, help drastically reduce nuisance
alarms and represent a truly cost-effective surveillance technology for
high-security applications.