When video analytics was introduced into the security market many years ago, the first phase was quite frustrating for end-users, installers and system integrators.
A lot of promises were made by software developers about how easily security can be improved by just adding some clever software to the existing security cameras. Plug-in tools were installed directly on the cameras or the software was run on the server side to analyse the data on a central system. In those early years, most projects failed, could not deliver and caused a lot of trouble within the industry. Too many false alarms, frustrated end-users and endless discussions about who is responsible for poor performances were the result.

The products were not robust enough, processing power at that time was mostly limited and the mindset of the industry was too self-centered. Integration and partnerships between companies were not standard.

As of 2018, the situation has drastically improved. Many projects and installations use all kinds of video analytics tools and most users are happy with the performance. Provided that a proper integration is done, the sky seems to be the limit. People counting, video fire detection, virtual tripwire, face recognition, heat maps, advanced searches and other applications finally made the life of security officers and business managers easier.

Now the question is, are we ready for the next round of innovation? Are you ready for Artificial Intelligence (AI)? Are you ready to create an intelligent world? In issue four of 2017, we gave you an insight on the AI vision of the big Chinese players, Hikvision and Dahua. If you have missed it, you can look it up on www.GIT-SECURITY.com/topstories. Creating an intelligent world was not only a big topic at the Milestone MIPS event in Barcelona and Las Vegas in February, but AI was also in the centre of a lot of presentations at Intersec Dubai.

When you look at the history of video analytics in our industry, one may say that talking about AI is too much of a marketing talk and it will take years until we see AI and Deep Learning applied in our industry. I don’t think it will take that long. Things have changed dramatically since the launch of video analytics. The biggest difference is that large companies with extensive resources in research and development have taken the lead here, in contrast to the small software start-up companies as seen in previous years. Users will have the choice to select between proprietary systems from one vendor with the advantage of the integration work done by another company or best-of-breed systems that combine the strength of all companies involved.

The mindset in the industry has changed, too. More and more companies develop systems together, partner with IT companies and experts from other industries. Some of the big players in security can rely on technologies they use in other parts of the industry like Bosch, who use sensors and algorithms from their automotive sector for applications within the security industry.

If there is one factor that may slow down AI development within the next years, I think it will be data protection initiatives especially in those European countries that have strict regulations or that have privacy protection as a big topic. 

No solution and no project can be developed without state-of-the-art products. Since many years we support product innovation with the GIT SECURITY AWARD. The award is going into the next round now and is a great opportunity for the industry to introduce products to an international audience and to test how popular their products are. Our jury already voted and the finalists will be announced soon. You can then vote for your favorite product of the year.