Advanced video analytics adoption rising rapidly, finds new NW Security UK study

Four out of every ten (41%) of England-based medium and large-sized businesses which are running CCTV systems have already deployed facial recognition analytics in their systems to capture human faces and compare images to human face databases to identify matches for access control, event security, or for public safety purposes.

One in six (16%) of CCTV system owners admitted to having access to this capability on their system but not yet going live with it.

Over a third of CCTV system owners (36%) in businesses with over 50 employees had already deployed some event or behavioural recognition analytics. Examples given included spotting loitering or highlighting abandoned bags (bags with no person nearby). A further 22% believed that they had this capability in their systems but had not yet turned it on.

ANPR and VMD

34% of CCTV system owners had deployedANPR to capture number plates at perimeter barriers

Just over a third (34%) of CCTV system owners questioned had already deployed Automatic Number Plate Recognition (ANPR) to capture number plates at perimeter barriers for example and the same number of system owners (34%) had deployed Video Motion Detection (VMD) to help reduce their system’s video storage requirements by only recording when motion is detected in front of a camera.

One in five (20%) claimed to have access to ANPR analytics in their systems but not to have turned it on yet. A further 27% claimed to have VMD capability in their systems which they have not yet activated.

Objection detection

Exactly a third (33.33%) of CCTV system owners in England claimed to have deployed object tracking which is a relatively new capability that enables security teams to track individuals from camera to camera through a large site in ‘auto track’ mode.

Nearly a third (32%) had deployed Objection Detection or Object Classification to help the system distinguish between humans, vehicles, animals, swaying trees, shadows, rain, luggage, water, roads, etc. 23% confirmed that they had this capability in their systems but had not yet turned it on.

Only marginally less (31.6%) claimed to have already deployed Directional Detection analytics to detect which direction an object or person is moving over a line. A further 28% claimed to have this capability at their disposal but not to have turned it on.

Optical character recognition analytics

Optical Character Recognition (OCR) analytics heavily used to read the identification numbers on parcels and other goods in transit has been deployed by 31% of England’s businesses in NW Security’s recent business survey.

A further 35% claimed that they already had this capability in their systems but had not yet put it to work.

Business intelligence-led video analytics

Heat mapping is used to detect crowds forming before events or analyse the busiest areas of a shop

Business intelligence-led video analytics was not far behind in terms of adoption. For example, heat mapping which is commonly used to detect crowds forming before events or analyse the busiest areas of a shop has already been deployed by 28% of CCTV system owners. Nearly another quarter (23%) claimed to have this capability at their disposal but had not yet configured it or made it live.

28% had turned on people counting analytics on their systems. NW Security discovered some of these people had adopted this capability to monitor room capacity levels for COVID Safety reasons. Nearly as many, 26%, though they had people counting analytics available in their system but had not yet made it live.

Facial and traffic detection analytics

Over a quarter (27%) recorded that they were using Facial Detection analytics and a further quarter (25%) had deployed traffic monitoring analytics in their systems. While 22% of system owners recorded making crowd density analytics live on their systems.

Even higher numbers (24%) confirmed they had access to crowd density measurement analytics but had not yet deployed it.

Adoption of video analytics

Frank Crouwel, Managing Director of NW Security, commented, “We have been surprised by the level of adoption of even fairly sophisticated video analytics adoption across CCTV system owners. That said, more and more camera vendors are offering analytics at the edge.” 

“Many video analytics types are present in over half of existing systems and 7 different types of analytics have already seen over 30% adoption across our total base of over 152 CCTV system owners of medium and large-sized businesses across England.”

An increasing number of new cameras sold feature more advanced analytics like event or behaviour recognition"

Advanced analytics in cameras

Josh Woodhouse, Lead Analyst at Novaira Insights, a UK-based video surveillance market research company and publisher of the ‘World Market for Video Surveillance Hardware and Software’ market report earlier this year, added, “Basic features like motion detection and virtual tripwire have been embedded in most cameras for many years. However, it is an ongoing trend where more analytics workload can be achieved at the edge."

Josh Woodhouse at Novaira Insights added, “An increasing number of new cameras sold feature more advanced analytics like event or behaviour recognition.”

“It is estimated that 43% of all professional-grade network security cameras shipped in the world in 2020 featured these more advanced analytics. This is forecast to nearly double to 81% by 2025, leading to cameras having the functionality for advanced analytics such as object detection or facial recognition as standard capabilities.”