With the introduction of 360-degree fisheye cameras with built-in “track-‘n-trace” capability in co?-mbination with video analytics applied to recorded footage, the retail sector has discovered that in-store surveillance can do much more than merely finding, tracking and identifying shoplifters, by analyzing and enhancing the shopper’s experience.
With the introduction of 360-degree fisheye cameras with built-in “track-‘n-trace” capability in combination with video analytics applied to recorded footage, the retail sector has discovered that in-store surveillance can do much more than merely finding, tracking and identifying shoplifters, by analyzing and enhancing the shopper's experience.
It is no surprise that large supermarket chains, department stores and others who offer loyalty and bonus cards use the data from receipts to analyze shopping patterns. Not only do they calculate your bonus points accordingly, they also tailor special offers on certain items based on your previous purchase. What they have not been able to — until now — is analyze, for example, which products you also looked at but did not buy, or how many times you visited a store without buying anything at all.
The stumbling block has been the relative inability to work efficiently with “big data” — such as surveillance footage — in terms of analyzing it for business intelligence purposes. But with an efficient marriage of sophisticated 360-degree technology and video data management, AMG-Panogenics and i3 have now demonstrated that a new world of applications are available to the benefit of many industry sectors.
With its long tradition for using business intelligence to analyze shopping patterns, average basket value and more, retailers are now fast discovering that a 360-degree in-store video surveillance solution can help deliver a wealth of additional information. Until now, in-store video has predominantly been used for finding, tracking, identifying and documenting (potential) shoplifters, but the leap to using the footage for analyzing behavior and movement patterns, performance monitoring at PoS, customers waiting in lines for service and much more is now in full progress.
Turning 360
David Meyers, CTO of AMG-Panogenics, explained the camera technology which facilitates multiple ways of using and analyzing video footage, and how this helps increase the overall ROI. “It was always our intention to bring to market a product range with technical innovations that offer the users real benefits. We provide up to 14 live, simultaneous, panoramic and digital PTZ streams while still recording the full scene at a rate of 12.5 fps. A certain amount of zoom is then available on playback; up to 10x optical zoom equivalent, from fully zoomed-out, when compared with SD cameras. All you need for live viewing is a standard Web browser. We also designed the cameras to work on open-platform software which makes it very easy to integrate into third-party NVRs, through our SDK. Onboard dewarping is another feature which we have found that customers are very keen on. It reduces network bandwidth and image processing loads on NVRs and promises much better storage optimization. When correctly placed, 360-degree cameras can significantly reduce camera counts and transmission infrastructure. No moving parts also mean no maintenance costs.”
Track-‘n-Trace in Practice
“The latest feature track-‘n-trace' detects and tracks motion in preset areas; our tracking camera streams can follow persons around within the 360-degree view of the camera. We currently have up to two tracking streams in the camera, and the main intention of the feature is to act as a deterrent to would-be shoplifters by displaying them on the in-store screen and tracking them as if they were being followed. Usually, surveillance cameras are seen only as an after-the-fact forensic tool; however, PanoCam360 provides an active loss prevention feature at the same time. The monitor can show both tracking streams in separate windows while the NVR is also recording the entire 360-degree scene from the same camera. We can work with up to 10 different views or area sectors with different ‘trigger points,' which we can predetermine. The triggers are set by simply pointing and clicking to define the relevant area. Once the area is entered by the shopper, the camera is triggered, and the tracking automatically starts.”
Footage Mining for Business Intelligence
Footage analysis is where i3 comes in, providing advanced VCA for users to get the most information out of their video. The software has now been integrated with the PanoCam360 and Vy Hoang, i3's Executive VP of Sales and Marketing, is excited about the application of the integrated solution in the retail sector. “Our CMS is a centralized server that delivers the ability to collect important customer data from the security video system and then combine with operational data in order to push real-time business intelligence to employees within an organization. The data is delivered in the form of reports that quickly identify lost opportunities, operational inefficiencies and areas of risk, without waiting for trends to appear. The fact that we deliver data in real time means that there is a possibility to apply corrective measures instantly, often for the benefit of better profitability. Say, for example, that you have a display of expensive sunglasses as a shop-in-shop function. It may be located adjacent to the cosmetics department or next to a custom jewellery counter. Not only do you want to ensure that visitors to the store area do not help themselves to a pair of shades from a loss prevention point of view, you may also want to track and analyze which brands and designs attract most customers, as this could prompt you to rearrange the display. There is a lot a psychology involved in how goods are presented for optimizing the sales and small changes in the display and use of the real estate can generate immediate impact on the bottom line. Further, you may want to track customers who bought sunglasses around the adjacent shop area to identify which goods they also sampled or bought in cosmetics and jewellery, thus investigating if certain behavior or movement patterns apply to certain categories of customers. Reports will include information on view time in front of a display, walking patterns within and around the area, as well as traffic counts within the area.”
In-store video footage can be analyzed in real time to provide valuable business intelligence and key performance indicators.
It is, however, important to note that traffic counters and analytics alone do not automatically increase sales. The information provided must drive strategic decisions and actions, Hoang stressed.
Focus
“It is also important to be aware that there are things you can actively influence and control, and others which you will not be able to influence,” Hoang said. “Noncontrollable events include things like weather, lunch groups, family groups, church rush, free store delivery (pick up at store from online ordering), unscheduled trucks and customer ‘loitering,' which can all inflict impact on numbers. Things you can control and work with to improve the customer experience include managing staff breaks, scheduling for coverage, clearing the DVR/NVR entry zone (removing items from an area that might affect the accuracy of people count or region count), managing cart return, managing morning sidewalk setup, customers using exit to enter, multiple trips for shop opportunities, manager-on-duty performance, evaluating associate performance and so on.”
Knowing how long customers look at a particular display can help decide where to place goods in the real estate or prompt changes to the way goods are displayed and highlighted.
The sample report shows overall customer traffic, comparing two different outlets in different geographical regions
Touch and Go
“Looking into the future, one of the immediate new applications — especially in the U.S. — is the requirement to have full-person image logging for identification and documentation of wireless/mobile/NFC transactions,” Meyers said. “Retailers are looking for better and faster ways to process purchases. The next step will be a payment facility where the customer simply touches his/her credit card to the nearest payment ‘pad.' When we move into ‘touch-and-go' on credit cards, the criteria for being able to document that it is the rightful owner of the card who is using it will increase dramatically. The ability to spot fraudulent cards and identify theft will also be very significant.”
There is little doubt that the pleasurable concept of “retail therapy” is about to see a remarkable change, as retailers increasingly mine/mind the new business intelligence capabilities from in-store video capture.