Unusual Behaviour Detection

iSentry is an autonomous video-analysis system that uses AI algorithms to learn the
normal behaviour patterns of objects captured on a CCTV feed. When activity does not fit
the normal pattern, the system triggers a series of human and automated responses.


Rule-based video analytics has strict requirements for a scene’s background, meaning that
the accuracy of analysis in varied and high-activity scenarios is limited and patchy. The iSentry
platform, on the other hand, uses AI combined with deep-learning filters to isolate and
differentiate between objects in difficult environments with remarkable efficiency.

iSentry views 100% of the feed but has deep-learning filters to suppress alarms for non-threatening objects and activity, presenting only relevant parts to the operator. This can reduce the amount of video an operator needs to interrogate by as much as 99%.

The iSentry platform substantially increases the number of cameras an individual operator can monitor — by between 10 and 100 times, depending on the environment — by presenting the relevant camera to the operator when the technology detects unusual behaviour.

iSentry is easy to implement and, because it is autonomous, simple to manage, with
no complicated administration requirements. iSentry can be deployed on existing
CCTV infrastructure — no need to ‘rip and replace’.

iSentry Capabilities:

Click below to find out more about iSentry capabilities.

Unusual Behaviour

Video Trip

Reliable indoor and outdoor
left-object detection

Application Arenas

Video-analytic software has radically transformed the potential uses of surveillance
technology. The number and type of functions have become dramatically deeper and more
varied. These are some of the key application arenas in the local and regional context.