Game Development Reference
In-Depth Information
further. In the past, one staff member viewed one camera on a single monitor. Now,
it is common to find them looking at scores of cameras linked to a single display.
After a few minutes of surveying, the human attention to video detail degenerates
to an unacceptable level, then the video surveillance becomes meaningless. The
problem in traditional surveillance video system is that although a massive amount
of videos are being recorded, but never watched or reviewed, due to lack of time
and the inefficient work by human. As a result, events and activities are missed, and
suspicious behaviors are not noticed in time to prevent incidents. Obviously, there
exists an enormous and growing gap between the amount of video data continuously
collected by cameras and our ability to efficiently transmit, store, and intelligently
analyze and digest visual information. Traditional video surveillance can no longer
meet the increasing demands of the industry.
Driven by awide range of applications in surveillance video, Intelligent Video (IV)
has been proposed and substantially growing in the past decade (Liu et al.
2012
).
Elliott defines the intelligent video system (IVS) as any video surveillance solu-
tion that utilizes technology to automatically, without human intervention, process,
manipulate and/or perform actions to either the live or stored video images (Elliott
2010
). Intelligent video system helps people to transform video surveillance into a
real-time, proactive, event-driven process by reducing the vast amount of informa-
tion contained in video and extracting the interesting semanteme information from
them. IVS includes many intelligent analysis modules, e.g., video motion detection,
camera tampering detection, people counting, face recognition, vehicle license plate
recognition, and so on. These are considerable tedious tasks for people to analyze
a recorded video by monitoring multichannel division screens for 24h. In addition,
intelligent video systems not only provide real-time abnormal event detection by ana-
lyzing images acquired from cameras, but also acquire continuous video sequences
from adjacent cameras using panning, tilting, and zooming (PTZ) with multichannel,
multiarea, and an immediate acquisition of the monitoring area video with complex
and vast environment (Nam et al.
2012
), which needs lots of complex computation
and is almost impossible for people to finish in real time.
Therefore, Intelligent video system is very important for future video surveillance
system and plays more and more important role in protecting us from harming and
facilitating our daily life. Some advantages of IVS are summarized in following
aspects.
Early warning for abnormal incident
. It can inform operators of abnormal
incidents, e.g., people moves in restricted area cars due to driving the wrong way or
attempts to tamper with the video surveillance cameras, which makes the operators
get rid of watching many monitors for long hours to spot suspicious activities. This
can significantly improve the efficient use of manpower.
Fast retrieval for special objects or events
. IVS can speed up the retrieval
of stored videos for special objects and events. For example, an intelligent video
system that has tagged the video stream with appropriate labels during recording can
automatically search through days of stored video to find the right video footage in
a matter of seconds. In the London underground bombing case in 2005, the suspect
was identified after inspecting two weeks of surveillance video captured from the
