The digitization of video technologies has opened a plethora of opportunities for using it in ways that were perhaps conceivable but wholly impractical back in the days of analog. Indeed, the limitations imposed by analog—not least the basic requirement for humans to be actively involved in the watching, reviewing, searching, and interpreting of images—meant that its scope was often extremely limited. Thankfully, digital video technologies are now readily available, and they have brought significant new opportunities, not least through the development of video analytics.
Understanding Video Analytics
Video analytics is a term that is increasingly used but it is important to begin to understand what it actually means and how it differs from other forms of video technologies and developments. The ECR Retail Loss Group recently undertook some research to develop a better understanding of how video analytics are currently being used in the retail environment including putting together a working definition of the term ‘video analytics.’ It was described as: a video system that uses the capacity of computers to automatically interpret digital images in order to generate clearly defined and actionable outcomes. There are three important points to note in this definition:
- The process is done by computers;
- It is done automatically—no human input is required; and
- It generates an output that is clearly defined and actionable.
As will be seen in the examples below, what this looks like in action is typically an alert when something happens outside of an agreed parameter, or an action is enabled if an agreed parameter has been met. So, for example, an alert is sent to a designated guardian when a person is identified entering a restricted area, or a vehicle is allowed to enter a restricted space because its license plate is on a designated list.
Current Uses of Video Analytics in Retailing
The same ECR report also presented data on the top five most deployed video analytics in retailing. Based upon responses from 81 retailers from 20 countries and a combined turnover of over €2 trillion ($2.38 trillion USD), the study found that four of the five were focused upon generating alerts when there was unauthorized movement or access to restricted areas (as seen in table one below).
In many respects, the popularity of this use case is understandable—using humans to keep watch over potentially hundreds of locations, looking out for the relatively rare occasions when a breach of security occurs, is inevitably laborious for the observer and highly prone to important events being missed. What this type of video analytic offers is a way to bring the vital few events rather than the trivial many to the attention of the watcher—in effect a digital filter.
The other most highly used analytic was the use of video to automatically control vehicle access to retail distribution sites. Through the use of automatic license plate recognition technology, together with a database of vehicles predetermined as having agreed access (sometimes within specified timeframes), retail companies are able to reduce the costs of managing access to some of their most valuable and vulnerable locations.
Future Use Cases of Video Analytics
The ECR research went on to explore what video analytics retailers were currently trialing or planning to use in the near future. Table two below presents the data for all the retailers taking part (most were grocery retailers) and as can been seen, the future focus is very much upon the checkout area in retail stores, in particular self-checkouts.
In other research, the ECR Retail Loss Group has highlighted the vulnerabilities presented by self-checkout systems not least in terms of customers not scanning items, misrepresenting cheaper items for more expensive ones, and walking away from self-checkout transactions without paying. For all of these problems it can be seen that retailers are exploring video analytics as a way to try and control them.
For non-grocery retailers a broader range of use cases is evident although the checkout remains a particular point of interest. As can be seen in table three, nearly one-third of respondents are exploring ways in which video analytics can now be used to try and identify suspicious behavior in the shopping aisle—in particular thieves that may be in the process of stealing produce.
Getting the Most Out of Video Analytics
The ECR research on the use of video analytics provides a fascinating insight into how retailers are currently using video analytics and what the likely areas of future development might be. It would seem an exciting and rapidly developing area, but, like any emerging new technology, it will be important to ensure that it offers value-added deliverables. Most video analytics operating in a retail environment are highly influenced by the complexity of the environment within which they are tasked to operate—the greater the complexity, the greater the challenge to get them to work as designed. It will be important, therefore, that retailers ensure any given video analytic genuinely delivers value rather than simply becoming an ongoing and unwelcome distraction for those tasked with meeting the core goals of the business.
Video Watch is a monthly column written by Professor Adrian Beck sharing insights on the proactive use and impact of video technologies in retail. It reflects the latest research and monthly discussions of the Video Working Group of ECR Retail Loss, the leading global think tank on retail loss. The research commissioned by ECR Retail Loss is supported by independent research grants provided by Genetec and other leaders in retail loss prevention.