Tim Compston, Features Editor at SecurityNewsDesk and SecurityMiddleEast.com, interviews Malay Kundu, Founder and CEO of StopLift Checkout Vision Systems, about the way that video analytics is helping to secure retail checkout areas from criminal activity.
As retailers seek to ring the changes at checkout areas unfortunately this is opening up new loss prevention challenges. One potential solution, which is now gaining traction in the Middle East and across the globe, to keep criminal activity in check is to refine video analytics even further to the point where it can flag up specific behaviour around scan avoidance, at manned and self-checkout areas, with a high degree of accuracy.
From Harvard project to the shop floor
Interestingly, Malay Kundu, from StopLift Checkout Vision Systems - the leading player in this field - says that the solutions StopLift has rolled out have benefited from his background in intelligent video content analysis. In fact Malay cut his teeth in facial recognition systems in airports and says that StopLift itself actually grew out of a Harvard Business School research study 'Project StopLift' on 'Retail Loss Prevention'. This was an effort which he reckons has stood him in good stead for the challenges found in today’s retail environment: “We are applying a lot of really hard-core high tech here. That is how we are able to detect ‘sweet-hearting’ [where there is collusion between employees and customers] and generally scan avoidance. We are doing a ton of this kind of stuff for manned checkouts as well as self-checkouts.”
Demonstrating just how far the potential to detect criminal actions, and errors, around the retail point of sale area has come is the software-based checkout vision system pioneered by StopLift. Malay says that the state-of-the-art ScanItAll solution automatically analyses video surveillance footage to interpret the behaviour of a cashier and customer and, crucially, the handling of individual items moving through the checkout area. By mathematically analysing the pixels of digitised video it can be determined if an item has been properly scanned. Apparently this approach is capable of ‘understanding’ a full set of fraudulent behaviours including, for example, when a cashier covers up a bar code by hand or purposely misaligns the scanner and an item. ScanItAll is also claimed to be very effective at detecting refund fraud, basket-based loss, issues at self-checkouts and other operational errors.
The scale of the rollout is certainly impressive with Malay confirming that at last count (April 2016) StopLift’s solutions had already detected and confirmed more than 1.5 million incidents. Many big name retail chains have now installed StopLift Checkout Vision Systems’ Scan-It-All video recognition technology at both manned and self-checkouts. With regards to the Middle East, Malay confirms that a key retailer in the GCC (Gulf Cooperation Council) area has already taken up the solution for their traditional checkouts, with a cashier set-up, which he reports is working well and hopes that more will follow.
On the self-checkout side of things Malay says that these are tending to be adopted in parts of the world, like the US and Europe, where shop workers’ wages are relatively high. He goes on to say that the latest innovation from StopLift - called StopLift Accelerator - is reportedly able to detect and notify a store about self-checkout theft in real-time, and, even distinguish whether it is someone's purse or mobile phone on a conveyor belt rather than a stolen item. From his experience talking to retailers who are putting in self-checkouts Malay reckons that they are certainly not blind to the challenges: "They are being very deliberate about testing these [self- checkouts] in their stores and making sure that staff know how to attend to them and that their customers are able to have a good experience. At the same time as doing that we are finding now that they are reaching out to us. They have realised that an inventory shrinkage increase, unfortunately, does often accompany putting in self-checkouts."
Drilling down to some of the practical issues retailers are experiencing, according to Malay, self-checkouts which don't have video-enhanced security in place are very prone to weightless theft: "Weightless theft is where the customer basically avoids the weight sensor. There are a whole variety of ways they can do that. They can just leave the items in the cart. They will sometimes pretend to scan an item but as they put it in the bag they will actually pick-up the bag so it never touches the weight scale. The weight sensor, understandably, is blind to anything which doesn't touch it." For Malay this stacks-up as the number one issue from an inventory shrinkage standpoint.
To put the vulnerability of self-checkouts, into some sort of context, Malay rounds of our discussion by revealing that StopLift's technology has found shoplifting to be as much as five times more likely to happen in the self-checkout lane - a dramatic difference: "If people don't have an enhanced security solution, like ours, they are opening themselves up to that sort of inventory shrinkage.”