The following is a short guide to AI CCTV systems, providing specific ‘how to’ details around specific artificial intelligence features - like how AI cameras can identify between humans and animals, and how ‘Smart Search’ works.

How to set up Virtual ‘Tripwires'

Virtual Tripwires allow you to detect movement via your CCTV system. Tripwires can be set up to detect specific direction of movement (e.g. A to B, or B to A) and also set to detect specific object types (e.g. human or vehicle).

How to set up Object missing detection

Object missing detection in CCTV allows you to detect whether an object has been moved or not, by drawing boxes around the specific object.

A minimum ‘missing’ time of 6 seconds or more needs to be set to minimise false alarms (e.g. the case where another object appears in front of the object), so objects cannot ‘immediately’ be detected as missing.

Objects also need to be of a significant enough size, and differentiated colour from the surrounding area to accurately detect the movement of the object.

How Smart Search works

Smart Search allows you to quickly and easily search for specific ‘events’ from your CCTV (e.g. a tripwire being activated) over a specific period of time. Smart Search means that you don’t need to watch hours of footage to see, for example, what happened at your property over a weekend.

Full Colour night vision explained

Full colour night vision means that the picture shown on your CCTV in low-light conditions is ‘full colour’ rather than ‘black and white’ (the norm for most CCTV cameras).

CCTV cameras with 'Full Colour Night Vision’ achieve this via a mix of clever colour-balancing algorithms and a bright LED illuminator. The Illuminator provides a level of light for the camera to operate in, and the algorithms help enhance the scene - providing a full colour image.

Full colour night vision is useful for helping enhance the performance of ‘detection algorithms’ - giving a higher detection accuracy.

Humans vs Animals algorithm (false alarm minimisation)

One of the biggest historic issues with external detection is the high level of false-alarms linked to animals and weather (e.g. high winds).

Using the latest technologies, high-quality AI algorithms can learn the difference between a human and animal (i.e. the shape and the way they move), minimising false-alarms. These same AI algorithms can also tell the difference between vehicles and humans - meaning you can, for example, set-up the system to only detect car movement.