Turning video into actionable intelligence

 

JOHANNESBURG, South Africa –  The role of the camera has changed completely, moving away from traditional security usage only, to become a multi-purpose, centralised device that can help organisations deal with operational, manufacturing and health and safety challenges, amongst others.

This is according to Stephanie Rosenmayer, business unit manager at high performing and secure ICT solutions provider, Datacentrix, who explains that there have been four major trends unfolding within the surveillance space over the past year.

Artificial intelligence (AI) on the edge

The first trend, according to Rosenmayer, is the slow but steady progression of AI – which started off as a centralised solution – to the edge. “We’ve certainly seen a recent uptick in the number of original equipment manufacturers (OEMs) that have developed AI solutions specifically linked to smart cameras that can be deployed at the edge. “This has had a major impact on the way that the edge is architected.”

The AI explosion

The next trend referenced by Rosenmayer is the continued explosion of AI. “A serious driver for the Open Security & Safety Alliance (OSSA), a non-profit, non-stock corporation working to create a framework outlining a common standardised platform for security and safety solutions accessible for everyone, is the development of an open source operating system for the safety and security industry.

“Some of you may recall the fate of BlackBerry-specific applications when standard Android and Apple platforms were developed, and consumers suddenly had access to thousands and thousands of new apps.

“We’re expecting exactly the same type of scenario for surveillance and AI. Once an open platform has been established, we believe that the barriers to entry will disappear, opening up countless applications that will run independently of the manufacturers.”

For business purposes, states Rosenmayer, this will be incredibly important, as organisations that have already made the capital investments to establish a physical infrastructure will be able to ‘chop and change’ the apps running on top of it, enabling customisation for specific environments, or even specific areas within the company.

The move to the cloud

“Several of the larger cloud providers have been eyeing the surveillance sector, and while this move is still in its infancy, there are a number of commercial models currently being tested in the market. However, there is quite a bit of work still to be done before we have mature commercial models that can be used.”

 Security and IT convergence

Finally, she adds, the convergence of security and the IT industry has become a reality. “Historically, security was a separate division often situated within the facilities side. Today, we’re seeing much more engagement between IT and security, addressing endpoint security, vulnerabilities, networking, bandwidth optimisation, the evaluation of which parts of the workload to move to the cloud, and more. It is anticipated that this organisational model will change over time.”

 Important IoT decisions to be made

Many companies, says Rosenmayer, have made a massive capital investment in terms of establishing networks, storage and compute. “They’ve outlaid spend on devices, access control, fire detection and many other components that need to ultimately integrate into a video/data management system. In fact, in most of the implementations Datacentrix has done, we’ve seen that this ‘infrastructure readiness’ portion of the surveillance portfolio makes up approximately 80 percent of solution deployment costs, while the application layer represents only 20 percent, despite having the potential to generate 80 percent of the value going forward.”

When looking at IoT, she explains, the decision around a video data management system will be the most important one a business will make for the next five years.

“Here, integration is key – and no longer just with cameras. Now, we need to look at the integration of IoT devices, with a specific view to creating meta data. Why is this important? The meta data generated by devices can be extracted for analysis and turned into actionable intelligence. So, we’re able to take the same camera, but generate valuable information for sales, marketing, production, operations and more.”