Sony SNC-CS50P - Manual (Page 8)

8
To understand the “Intelligent Motion Detection” and
“Intelligent Object Detection” functions built into the
Sony SNC-RX550/RZ50/CS50 Series of network cameras, it
is important to first understand motion detection in
general.
What is Motion Detection?
Motion detection is a relatively common feature built-into
surveillance equipment.
One benefit of motion detection is that it can greatly
reduce the required storage capacity of a recorder.
A surveillance system can be configured in several
different ways depending on what is being monitored.
For example, the recorder can be set up to store low-
resolution images at a low frame rate or to record nothing
at all to save storage capacity under normal conditions.
When an alarm is triggered by movement, the recorder
automatically begins to record higher-resolution images at
higher frame rates so that critical scenes can be clearly
captured.
Another benefit of motion detection is that it can alert
operators when movement has been detected in several
ways, for example by sending an e-mail notification,
providing an audible alert with a pre-recorded audio file,
flashing an alarm message on the monitor, or by
displaying a full-screen image from the camera that
detected movement. What‘s more, the alarm can trigger
local image recording or move the camera to a preset
pan/tilt/zoom position to get a closer look at the
monitoring object. In addition, the system can be
configured to perform any of the following actions and
more when movement has been detected: sounding an
audible alarm, turning lights on/off, triggering a door
lock, etc.
Motion Detection Basics
Conventional Methods of Movement
Detection
A variety of detecting methods have been designed into
surveillance equipment from the time these systems were
first sold. Sony has incorporated movement detection
features not only in recorders but also in cameras.
Sony first-generation network cameras that employ the
JPEG compression format, such as the
SNC-RZ30/Z20/CS3, incorporate a basic detection method
that compares the average change in luminance levels
between adjacent frames (i.e. adjacent JPEG images) on a
pixel basis. If the result is greater than a preset threshold,
then it is treated as motion in the monitoring area, and
triggers an alarm (Fig.1).
Another detecting method that was incorporated in Sony
second-generation network cameras, such as the
SNC-RZ25/DF70/DF40/CS11/CS10, employing the MPEG-4
compression format, utilizes motion compensation inherent
in MPEG-4 compression. Motion compensation is based on
movement within 16 x 16 pixel areas of an image called
“macroblocks.” In the motion-compensation process,
motion vectors, which are based on the direction, speed,
and distance of a moving object within each macroblock,
are determined. These motion vectors are then added and
if the resultant vector exceeds a preset threshold level, an
alarm is triggered (Fig. 2).
Fig. 1 Detection Using Average Change in Luminance of Pixels
Previous image Current image
Previous image Current image
16 pixels
16 pixels
Motion Vector Motion Vector
Fig. 2 Detection Using Motion Vector