The Rise of the Visual Internet of Things
James Wickes, CEO - Cloudview
When I think about the Visual Internet of Things, I see a huge array of possibilities running through my mind’s eye, in which visual data plays a role in a whole host of smart, intelligent systems. And I also see an industry that hasn’t really woken up to the potential yet. That’s why we’re publishing our latest White Paper Visual IoT - Where the IoT, Cloud and Big Data Come Together.
Cloudview's research suggests there are 8.5 million cameras in the UK. Between them, these cameras create a massive 10.3 petabytes of visual data every hour. It’s a big number, and hard to imagine. But if we assume that the average MP3 is encoded at 1Mb a minute, and the average song lasts about four minutes, then a single petabyte of songs would provide more than 2,000 years of continuous play. So we’re talking a lot of information here.
Much of the visual data that’s recorded by these cameras is for surveillance purposes. But it is increasingly used to enhance Internet of Things applications, and it has vast potential to do so. In our White Paper we look in detail at some of the current ways in which visual data is helping the IoT be more effective, and consider some future possibilities. There are opportunities literally everywhere you look, from transportation to commerce, from smart cities to healthcare.
Central to implementing all these possibilities is the convergence of cloud-based storage, analytics and our increasing ability to handle masses of data, of different types, at speed. When automated systems can map the flow of people around museums, airports, shopping malls and city centres, we can start to better understand natural movement patterns, and then design taking these into account.
This works on the scale of a single individual too. When an automated system can understand the difference between someone moving around their home in the normal course of daily life, or moving awkwardly or being still for unusually long periods, it can help us manage care systems. Similarly, movement patterns can identify potential self-harmers on railway stations, or even potential acts of violence in public places.
Where do the millions of existing cameras fit into all this? In many instances the visual data they capture is ‘single use’, and once it’s been recorded it can sit in a file store, never being looked at. What if that data could be shared? It’s not going to be appropriate in all cases, of course, but where it is appropriate, it can be transformational. Consider how bringing cloud storage and remote access to the CCTV sector is reaping benefits already, and how it could enable much more.
For example, housing providers whose visual data is stored in the cloud are able to access that data anywhere, from any device. So they can take snapshots of work that needs doing and share these with contractors, and check the results of contractors work remotely. If necessary, they can also share live streams with first responders. That’s a far cry from recordings that are sent to an on-site digital video recorder that needs to be visited before any visual data can be accessed.
But let’s take that a little further, and imagine analytics being brought into this kind of system. Now the system can alert housing providers if contractors have not completed a task by the due date, it can identify for itself if bins have not been emptied, or stairwells not cleaned, and let the housing provider (and contractor) know. The system can also be asked to monitor specific front doors, and issue alerts if someone elderly or vulnerable has not left their home for a while. Using its predictive analytics, the system can even spot potential anti-social behaviour, so that problems can be dealt with before they happen rather than retrospectively.
Legal provisions around privacy protection are about to be overhauled as the UK’s Data Protection legislation gets replaced by the more sophisticated General Data Protection Regulation, which comes into force in May 2018. This is a forward-looking piece of legislation that caters for current and future technologies in a way the older Data Protection Act did not. It places requirements on those who handle visual data, and provides for the appropriate checks and balances to be put in place to protect personal privacy. We look at the evolution of data protection legislation, and the impact of the GDPR, in more detail in an earlier White Paper – Watching the Watchers.
Having appropriate legal protections in place is a key piece of the puzzle that will help us make better use of visual data. Along with cloud storage and increasingly sophisticated analytics, the future of visual data that is in my mind’s eye, is starting to become a reality. It is a future that moves us away from control and surveillance, and towards systems that use what they can ‘see’ to keep us informed – and help us better look after places and people.