iSentry learns normal behaviour by observation and detects unusual activity

Using advanced artificial intelligence software the iSentry program is able to learn:

The activity patterns of the things in the camera's view.

To filter out normal activity and detects unusual activity.

It learns by observing what happens normally. No teaching or special training is required.

iSentry will learn what it needs to know from the video footage it watches every day. The speed with which iSentry learns make it extremely easy to install or redeploy cameras.

iSentry can learn behaviour for different time zones and days so that iSentry learns the behaviour specific for that time or day.

For example, a crowd of people on a Saturday morning might be normal but a crowd of youths at midnight on Monday might be trouble.

A busy road may be slow in peak hour but flows well at othertimes. iSentry would learn that slow traffic at noon is abnormal and raise an alert.

In the event of an alert a number of different communication options are possible within the framework of the iSentry system.

As a default each alert will be brought to the attention of a human operator.

iSentry vs Video Motion Detection (VMD) systems

Most companies offer Video Motion Detection (VMD) systems. VMD is a poor method of filtering video. It causes too many false alarms and it often doesn't pick up real alert situations. Also, it is almost useless in busy environments where there is significant movement.

The iSentry motion analysis system offers capabilities far superior to any VMD system. This allows a far greater level of filtering to be performed on video streams, while even small unusual events are still easily detected and can be brought to the attention of an operator.

The iSentry program allows the activity in each camera view to be gradually learned, without need for human intervention to identify perimeters, hot-spots or allowable motion flows.

Visit FAQ on Ultra-Smart | Next Feature | Back to Product Overview