Video surveillance cameras are increasingly being regarded as smart IT devices which happen to be equipped with an imaging sensor and a lens. There is so much more you can do with cameras now that they have the capability to run on-board video analytic applications such as heat mapping, people counting and queue management, as well as pandemic related face mask detection, social distance measuring and occupancy monitoring solutions. With the processing power to do so, it makes sense to mine the captured information at the Edge and thus avoid transmitting large amounts of data over the network.

Whilst manufacturers such as Hanwha Techwin are supplying cameras pre-loaded with these applications, as well as Intelligent Video Analytics (IVA) such as tampering, directional detection, defocus detection, virtual line, enter/exit and motion detection, the Edge provides the perfect opportunity for specialist third-party software developers to innovate by way of developing ground-breaking, serverless solutions that meet the specific needs of individual vertical market sectors.

Cost saving and scalable solutions

As an example, there is a serverless ‘small site’ ANPR solution available which automatically controls the movement of whitelisted cars through barriers via camera relay outputs, as well as provide valuable car park management information, such as ‘time spent’ and occupancy rates. It does so without users having to incur the cost of installing and running the application on a server, as up to 4 cameras (1 master camera and 3 slaves) are able to simultaneously capture and transmit video analytics data to a convenient user interface.

Serverless edge-based solutions are scalable and thus provide users with the flexibility to incrementally expand their systems at any time, without having to procure an expensive server.

Deep learning AI video analytics

The recent introduction of affordable cameras supplied with Deep Learning AI video analytics onboard, has further enhanced the ability of cameras to be used as detection devices. This higher level of performance, using Deep Learning AI based video analytics means control room operators and security personnel can focus on responding to real incidents and emergencies and not waste time and effort on false alarms. In addition to extreme accuracy, deep learning also allows operators to search for specific features and attributes, including a person’s age group and gender, and whether they are wearing glasses, a hat, or carrying a bag.

Retail and traffic applications

Cameras equipped with Deep Learning AI are particularly suitable for applications which require a greater degree of sophistication than that offered by traditional video analytics. It enables, for example, retailers to capture and analyse business intelligence such as age and gender, allowing them to granularly analyse customer demographics and in doing so, gain a greater understanding of customer behaviour and buying patterns.

Road planners, traffic regulation enforcement authorities and police are also now able to take advantage of AI cameras to identify the make, model and colour of vehicles, in addition to recognising car number plates. The shortly to be launched edge-based Wisenet Road AI intelligent traffic management solution will, as an example, utilise Deep Learning AI video analytics to identify over 700 vehicle models manufactured across 70 brands. The data can be used to conduct surveys to gain a greater understanding of road usage, as well as help accurately identify vehicles involved in traffic infringements.

Opportunities for integrators

System integrators will find the Edge provides them with the opportunity to offer their end-user clients innovative and sustainable systems. In doing so, they will be able to capitalise on the latest technology to deliver solutions which would not have previously been thought possible, when low resolution, monochrome CCTV images had to be expensively transmitted and remotely monitored from a central location.