The surveillance market is forecast to reach $42bn by 2019. On the face of it, this might seem like a surprisingly high figure, but if you dig deeper and consider that businesses may need to hold their CCTV footage for weeks, months or even years.
This explains the boost the market is seeing, as CCTV systems need to be more sophisticated and scale much better to support these requirements. But it’s not just the sheer scale that will drive the burgeoning market forward; new innovations are being developed to enable businesses to use data analytics to transform the value they derive from their surveillance footage. Multiple use cases are emerging meaning that CCTV is no longer just required for security. It can enable businesses to boost the bottom line.
Currently, many businesses use surveillance technology in a fairly conventional way. Either people or simple software monitors what’s happening as it happens, or looks up video footage in the wake of an incident. Advances in data analytics, however, are opening up a new world of potential applications: CCTV footage can be used to prevent crime before it happens, enhance the efficiency of supply chain or logistics operations, or even improve shop layout by assessing consumer movements around a store.
For organisations to achieve these outcomes, three key things need to happen:
- Preparation for scale: The first challenge lies in finding a way to reduce the cost and increase the efficiency of storing the increasing amounts of digital information in a growing CCTV estate. Longer retention requirements, some driven by EU and local law, and others driven by internal policy, will also boost the need for a huge data repository that can grow effectively. The key lies in unlocking scale-out architecture, allowing you to expand the pools of storage resource available without finding your costs spiralling out of control.
- Implementation of data lake resources & automation: This will allow real-time analytics without incurring a loss in the fidelity of production IT systems, or an unnecessary burden in terms of infrastructure and full-time employee (FTE) support. Real-time analysis of any data is fairly resource intensive; trying to assess patterns from video content even more so. Without a data lake resource to support, it’s simply not possible to do data analysis on this scale.
- Creation of an analytics lab: As my colleague Seb Darrington has written, the outcomes of analytics programmes are not pre-determined – it’s important to establish a mechanism to create ‘data theories’ and explore or test the data available to see what it can tell us. It may not be obvious that examining video feeds could be a good way to determine the value of real-estate to retailers in airports, for example, or could help expose fraud in a supermarket, or drive real-time alerts to security guards on suspicious behaviour in a shopping centre. These theories needed to be tested and assessed by a team of professional data and video tinkerers, so creating a framework in which they can operate is vital.
CCTV’s capabilities are moving far beyond security and monitoring services: retailers are looking at how customers browse a store, informing decisions about product placement and shelve optimization to boost sales. Leading banks are using analytics to monitor cashiers cashing up at the end of the day, double checking procedures are being followed and limiting the risk of fraud.
Analytics has also transformed security monitoring whether it’s on public transport, in a retail store or shopping centre or on a high street: instead of a CCTV operator spotting someone stealing from a shop or taking someone’s bank card from a cash machine for reactive security, analytics can be used to detect unusual behaviour, tag suspicious characters based on their activity and alert security to take a closer look or intervene where necessary to stop an incident happening. This can range from personal theft right up to large scale terrorist attacks. This kind of intelligence isn’t just the thing of CIA TV programmes like Homeland (Season 4 spoilers at link); it is rapidly shifting from fiction to reality.
Have you seen or deployed any interesting video analytics applications? Do you think your organisation is well-equipped to cope with and capitalise on the explosion in CCTV? I’d love to hear your thoughts in the comments.