Tag Archives: retail video solutions

Data overload, or information you can use? | Enter Data Visualization

Security and video surveillance systems can tell us a lot about our business – when the highest number of employees come through the front door, which way they go after they enter the building, or how many times each day a specific door is opened.

The good news is that all of this information and much more – from access control and event reports, video surveillance footage, and associated analytics data – can be captured by various edge devicesand stored within the system. This wealth of data is ready to be analyzed and turned into actionable intelligence that can help to detect and prevent issues in operations, equipment, or policies.

The bad news is…well, it’s the same. Without a means to make sense of it all, this wealth of data can be overwhelming and unusable.

Retailers, for example, are now able to capture minute-by-minute video of people in their stores — moving through the aisles, standing in front of displays, and making purchases at the register. But all of this can just be more video in the recording system if it isn’t put to use.

A data visualization tool can take the vast pools of information — security’s Big Data, if you will — now at the disposal of security personnel and management and turn it into more comprehensible data such as graphs and charts that correspond to particular video clips or POS data.

Perhaps your goal as a security manager is to determine whether you need additional guard coverage within a medical facility. Even though you have a card-based access control system that tells you who is entering the building and when, you now want to look more closely at which areas of the building they are accessing so you can beef up patrols in the more highly trafficked areas.

By deploying people-counting analytics tied to your video system, you can target specific doors or hallways that are covered by cameras and record video each time someone uses that door or enters a hallway.

Data visualization can then take you to the next step, which is putting all those video clips into usable charts. You can sort the data by day, for instance, creating charts that show you days when the traffic patterns are the heaviest. But you can even drill down further, looking at the time of day when particular doors and hallways are most heavily used and chart that information as well. Once you have the data set you want, you can look at the corresponding videos to see the actual activity.

Data Visualization

The result may be that you see the area outside a newly relocated lab is generating a lot of traffic outside of regular business hours. That area may require additional checks by a security officer to monitor the activity.

Retail can definitely benefit from the conversion of information into data visualization. Not only can retailers monitor traffic, just as the security officer did in the previous scenario, but by adding in the POS data, store operators can look at how people movement corresponds with sales.

Using data visualizer technology, it’s possible to create a data set that shows how many people walk by a handbag display, how long they looked at the products, and then how many actually made a purchase. All this is possible just by dragging and dropping the results into the data visualization software — sales from POS, plus those who lingered at the display for more than 10 seconds. Based on the results, it may influence the store owner to add sales staff at certain times of day or maybe relocate a display to a more highly traveled section of the store.

Put simply, data visualization is all about recognizing patterns and trends. Once you’re able to start thinking about all that data in those terms, it is less overwhelming and much more useful. And that really is the good news.

 

Are you already using Data Visualization software with your security footage? If so, in what ways? If not, any thoughts on how it could benefit your business? Please leave your response in the comments section below.

Turning up the heat on video analytics

For the average person watching people come and go from a retail store, a sporting event or even a hotel lobby, the movements can seem random, with no apparent pattern behind them.

But for those who are deploying the latest in video analytics, those same travels within a defined space can be tracked and analyzed to allow store operators, arena owners and hoteliers to make key business decisions related to merchandising, personnel positioning and, of course, security.

Heat mapping — one of the newest analytic tools available — provides a visual interpretation of traffic patterns. The movements of those people who appear to be merely walking through the store are now captured and analyzed, and the resulting data can be looked at more closely.

video analytics heat mappingThis becomes important in a retail setting as stores look to maximize the effectiveness of their displays and increase overall store performance. A promotion may seem successful on paper, but if no one can find the display within the store, or they don’t stop to really look at it, then sales will suffer. Heat mapping can show retailers the traffic patterns within the store and help them determine where the items should be placed.

Beyond displays, stores and venues can also use heat mapping to help them determine where to place permanent items such as ATMs. It may seem logical to put the machine at the entrance, but if the goal is to get people into the building, and heat mapping shows that people use the machine but then turn away, it may be better to place it within the facility.

Closely aligned with heat mapping is another analytics tool that helps make sense of the traffic analytics. Once it’s established where and how people are moving within a retail venue, the next move is to see how long they dwell or linger within the area, checking out the products on display.

Looking at the retail example, once the display is in its optimal location, the goal is to get shoppers to look at the products being sold and make a purchase. Dwell data tells us how long someone is stationary; linger does the same, but the person is usually milling about, rather than standing in one spot.

By integrating dwell and linger information with Point Of Sale data, it is possible to determine how many people stopped in front of the display for several seconds, or moved within the general area, are converted to buyers.

Dwell and linger data can also be used to determine how to assign personnel within an area. If analytics show that people are milling in a specific spot within a store, it may be necessary to send over a salesperson to offer assistance or, in the case of hotel, an additional desk clerk. Analytics that show a group gathering for a pre-determined period of time could also be a sign that security needs to check out what is going on as it could indicate suspicious behavior, such as a shelf sweep.

Of course, all this information won’t be of much good if it isn’t easy for the system’s user to retrieve and analyze the data. While security personnel are familiar with analytics, it may be new to those on the operations side. But today’s analytics are tailored in such as way that the information is accessible to everyone, no matter their IT expertise and no matter the size of the operation.

 

Do you have any questions about Heat Mapping? Please ask in the comments section below.

Fisheye cameras | Moving toward that perfect view

Having eyes in the back of your head — an attribute relegated to vigilant parents, strict librarians, and more than a few teachers — is a condition that would serve the security industry well. After all, who doesn’t want the ability to have eyes on the action, wherever it’s happening?

Since such forms of sight are a physical impossibility, security personnel have turned to cameras to be their all-seeing counterparts. The problem, however, is that even cameras don’t always capture every movement and moment in a critical scenario. Many cameras have fixed views and pan-tilt-zooms, and even with their wider range, can sometimes be looking in the wrong direction when they really need to be focused elsewhere.

Fisheye CamerasThus, the industry developed the fisheye technology or 360-degree cameras to provide that all-seeing viewpoint. As camera resolutions increase, the fisheye is turning out to be an especially advantageous choice for situational awareness within the retail industry, not only because these cameras fit their coverage needs, but also because stores can now lower their camera count and save a bit of money as well.

That’s not to say that the fisheye is limited to retail settings. These cameras, with their sweeping views (for example, 360 degrees if ceiling mounted or 180 degrees if wall mounted), are also ideal for university lecture halls, elevator banks, indoor parking areas, hospitals and casinos. As with all camera installations, it is important that camera selection is based on the combination of surveillance system needs and camera capabilities. For 360° cameras, the resolution “sweet-spot” for a 5MP fisheye capturing sufficient details is 20-25 feet. Due to the extreme wide angle lens, objects in the camera’s field of view beyond this distance will become too small to capture details. Thus, other types of cameras would better serve a large, outdoor parking lot application.

As with any component in a surveillance set up, the fisheye is at its best when used in concert with other cameras. For instance HD fixed cameras might be a better choice over the cash drawers, or multiple fisheyes could be used together to surround an HD PTZ unit that can zoom in and closely track events or suspicious individuals identified in the fisheye video. In some cases, however, a single fisheye can cover an entire room, such as classroom, without missing anything.

One challenge with the fisheye is the reality of image distortion at the edges, due to the ultra-wide field of view. Dewarping technology within the camera, as well as within the client, can address this problem and produce a corrected, flat view.

For security personnel, there can also be a learning curve as they become comfortable with navigating around the non-normalized or warped view. However, in an active surveillance situation, officers will want to work with the normalized view that can be produced on the edge by the camera or on the client-side with dewarping completed on the server.

The 360-degree fisheye camera expands the line of vision and takes surveillance into far corners not easily seen before. Over time, working with the views from this latest entrant into the video field will become second nature to surveillance system operators, and security personnel will relish having a new tool in their arsenal as they look to achieve all-seeing status.

 

Let me know other ways you are using Fisheye cameras by leaving a comment below.

Security Analytics Thwart Retail Theft, But Tell Us So Much More

For one department store, it seemed like business as usual as shoppers perused the racks of merchandise, picking out the latest fashions. But in the security office, the person on duty was alerted to an anomaly — on a merchandising display, an entire section of high-priced jeans had been removed quickly. Was this an overzealous shopper looking to try on multiple pairs, or someone from an organized retail theft ring sweeping up inventory?

In this specific scenario, an officer was made aware of this potential incident through the deployment of video analytics as part of the company’s overall surveillance plan and was able to respond quickly. The retailer had established a set of rules within the system’s software so an alert would be issued if more than half the inventory on that rack was removed in less than one minute —  a common scenario known as “shelf sweep” when shoplifters are at work.

The same analytics that are deployed for high-risk security settings, such as airports and government buildings, are equally at home in a retail setting. Security Analytics Retail: Tyco Security ProductsLike the shelf-sweep rule, similar guidelines can be created so a security officer can be alerted when someone enters a storeroom after hours or when an object, like a package, is left in one place for too long.

But what is equally exciting is that, because of the nature of analytics and its information-gathering abilities, its applications can go well beyond the security realm and become a boon to other store personnel.

Retailers who have included analytics in their security systems to both detect incidents as they happen and aid in forensic investigations of thefts, slip and falls and other activities, are expanding the reach of this investment and applying it to merchandising, marketing and operations.

After all, these cameras are operating 24/7 so why not take this database of information and look at it in the aggregate?

Let’s go back to that department store and see how analytics can help sell some handbags. Data supplied by the point-of-sale system will tell the store operator how many designer purses have sold, but not how many potential sales of those handbags there were on a given day.

By using the video system, the store can track how many people came through the doors (the total pool of potential buyers), and then break it down even further, using rules within the analytics to narrow down how many people walked down the aisle where the handbags were merchandised and then how many of those shoppers lingered for more than five minutes at the display. This information, teamed with the POS data, can now give that store’s manager a conversion rate on the sale of her designer handbags.

Armed with the knowledge of how many bags were sold vs. how many people stopped to look at them, it may mean that the purses are in a great spot or, if the conversion rate is poor, this is an indicator that the bags need to be displayed elsewhere or the signage improved or the price reduced. Analytics won’t read the minds of the shoppers, but the data can provide a good snapshot of what occurred within the store. Using analytics to determine traffic numbers and patterns can aid in where to locate merchandise and even help set the number of checkouts needed on a given day.

From a security standpoint, analytics in video surveillance is a necessary part of doing business, but by expanding the potential of its use, the entire retail operation can benefit — deploying the same equipment, but just tweaking the data to fit each users’ needs.  It can be win-win for both security and operations, and who doesn’t like that?

For more on the role of video analytics as part of a retail security solution, download our recent white paper on Video Analytics in Retail.


Download The Video Analytics White Paper