Retail Fraud Gravy Train Could Be Derailed

With the new year of 2009 upon us, we're on the tail end of the holiday season and that means those customer service lines are packed with folks trying to return unwanted gifts for cash or store credit. However, what you may not know is that some of those customers are returning goods they never bought-- and cashiers may even be in on the tricks as well.

We blogged a few weeks ago about the growing trend of retail fraud in retail stores nationwide, and as we said, it's as simple as walking into a store, picking up an item and returning it for a full refund. Return fraud will cost retailers an estimated $11.8 billion in 2008 — $3.54 billion during the holiday season alone.

3VR was featured last week in a segment on CBS 5 San Francisco about our proactive approach to eliminating retail fraud with our surveillance software. The piece focused on our unique facial recognition and surveillance technology, which immediately sends alerts to security personnel the moment suspected fraudsters enter the store. Using the 3VR solution, retailers have access to everything they need to apprehend fraudsters before they can make their escape into a crowded mall or parking lot and before employees clock out for the night.

Research is showing that due in large part to the ailing economic conditions we're facing, retail thefts are no longer mostly limited to customers in urban areas. Both customer and employee hands are stickier this holiday season in more rural regions, demonstrated in the spike in crime in states such as Arizona, Colorado and New Mexico. Coupled with smaller budgets that have forced stores to cut prevention costs, retailers are stuck with tremendous financial losses. Organized Retail Crime (ORC), or "sophisticated crime rings that steal and stockpile huge quantities of merchandise that is sold later to unwitting buyers" in flea markets, pawn shop and other outlets, has also increased in these tough times.

"It's not pure need, but times are tough and the economy triggers people making bad decisions," said Joseph LaRocca, vice president of loss prevention for the National Retail Federation (NRF). An October NRF survey revealed 79 percent of retailers polled have experienced an increase in employee theft compared with the same period in the previous year, and according to LaRocca, retailers lost $34.8 billion in 2007 from employee thefts and shoplifting combined.

Ranging from simple shoplifting tricks and cash siphoning from registers to colluding with customers in sweethearting (when the store cashier works with the customer and intentionally undercharges by skipping items when checking out), retailers are fighting an uphill battle against employee fraud this year. As recent statistics and coverage is showing, proactive solutions like 3VR will be ever more critical in 2009 for retailers looking to prevent these criminal acts from both sides of the transaction and survive the slower revenue streams expected in the year ahead.

Merry Christmas Drivers!

Santas helpers came early this year with a special gift for Tempe, AZ Drivers:



Hopefully these gift-givers will visit San Francisco next. Merry Christmas!

Advertising Gets Face Lift

Times Square. Piccadilly Circus. Even the turnstiles of the MTA subways in Manhattan.

These places are all highly trafficked areas with little advertising space to spare, and in such congested areas, marketers have tremendous difficulty determining the types of billboards that draw the most attention from audiences.

New technology out of Tokyo offers a solution to better measure the billboard's effectiveness – by utilizing surveillance technology. Slated to debut in a Tokyo railway station in January, NTT Communications is creating a billboard that stares back.

According to IDG News Service, the technology is able to measure how many people look at the billboard in order to determine the ad's ability to catch a consumer's attention.

Mounted just above the billboard, a digital camera coupled with basic facial recognition software will scan crowds and track if a person is looking directly at the advertisement. A second camera will be mounted underneath the billboard to track general crowd size in proximity to the ad. For privacy pundits, NTT is fast to stress that the cameras do not retain personal information about those tracked.

Capital or lowercase letters? Bold or neon color? Video or static image? This innovative approach will better measure how style, location and size aspects affect consumer attention and is the next step in adapting surveillance technology for other purposes.

Dunkin' Donuts actually performed trials a few months back to use facial scanning technology to target advertising to your age, gender and demographic group. Noted by the author, this was the "the first time it has been used by a mainstream advertiser in the US and works in the same way as systems used by law enforcement and emigration agencies to spot criminals in crowds."

Like the Tokyo billboard set-up, Dunkin' Donuts used cameras mounted above a screen to capture a customer's face and analyze facial features such as eye distance, jawline and cheekbone structure with complex algorithms. This information was then used to select ads more attune to the person's characteristics – a process about 85% accurate in tests.

Both Dunkin Donuts and NTT Communications show how surveillance technology can be modified to work for sectors other than security. Whether it is the donut shop around the corne or the train stop nearby, organizations are taking a page from Minority Report to operate more efficiently.

Faces: The New Fingerprints?

As we learned from the International Biometric Group's recent report, fingerprint scanning is now the biometrics option of choice for law enforcement agencies.

But one sheriff's department in Washington State has gotten a jumpstart in taking things a bit further-- the Pierce County Sheriff's Department was the first in the country to try out the MorphoFace Investigate system, a facial recognition system by Sagem Morpho, and their new technology recently made its inaugural criminal catch.

This past fall, the department caught on to an alleged bank fraud ring, and they were able to break up an organized effort to steal ATM cards and make fraudulent bank withdrawals by doing a database search around the ATM photograph. Two potential mugshot matches came up in less than 5 seconds, and it turned out the two matches were the same person, who happened to have a documented history of identify theft.

Police were able to move quickly, get a search warrant, and find the evidence they needed to make their arrest. Posthaste.

What's critical to note here is that facial recognition technology is adding a new layer to the biometric tools already in place. With this layer, we have access to new levels of speed and accuracy, and as the Pierce County forensic investigations manager says, the facial recognition system "eliminates 80 to 85 percent of the work in booking repeat offenders."

Maybe this year Santa will be making his list and checking it just once?

Security in 3D: Coming Soon to an Airport Near You?

Security engineers left and right have been attempting to crack the code on technology to match CCTV photos of faces with image archives in order to capture known criminals and thieves. Recent research out of Arizona State and Michigan State show that 3D laser scans could be eventually utilized to alleviate problems with lighting and photos angles in public areas with fast moving crowds.


Using a newly developed program, Dirk Colbry from ASU and his MSU colleague, George Stockman, performed 300 laser scans of 111 different faces using a commercial scanner to store images, while a horizontal plane of laser light passed over the subject's face. These images were then manipulated to create 3D models of each person's face.

The results were superb-- with the new system, different scans of the same face were matched even when lighting was unusual or the angle from which the images were taken was off by as much as 30 degrees. The error rate was a startling one percent.

This research demonstrates a new step in the advancement of surveillance technology, although its implementation will need to overcome the high price, slow scan speed and short-range sensors that may serve as roadblocks. The current scanner price of $50,000 needs to drop to about a tenth of that price to encourage widespread deployment. Additionally, scans currently take 2 to 5 seconds, making the technique obsolete in large crowds and only functional at choke points, such as airport and train station security checkpoints, where passengers are forced to wait in queues and in close range of scanning devices.

Being able to accurately identify someone who has paused, short-range, at a checkpoint and presented thier face, or iris, or fingers, or hands to the confidence levels demonstrated here has been possible for quite some time. But sometimes it's hard for me to get too excited about most of those approaches. That's because what we would all like to get out of any new biometric systems that get deployed at the airport are shorter lines and fewer checkpoints not more!

At 3VR, we're constantly looking for new algorithms to improve our facial recognition analysis features to combat the disparities in expression, lighting and angle. But to date, we have limited ourselves to writing algorithims for use in conjunction with conventional video cameras. That's because CCTV, and even new IP camers, are cheap, prolific, and offer something that laser scanning systems can't, the ability to work at a distance with uncooperative subjects.

Though current facial surveillance approaches fall somewhat short of 99%+ accuracy of 3D laser scanners, I can say that some of the techniques we are pioneering today hold strong promise of closing that gap. By processing streams of facial data from standard CCTV video feeds, it's possible to create an extremely accurate facial model; maybe one that will someday rival 3D scans. When will facial surveillance catch up? I can't say exactly. But, I do know that technologies like what we deploying at 3VR will get there long before a $50,000 laser scanner becomes as cheap as a video camera.

Biometrics: Global Demand on the Rise

It allows you to breeze right through the grocery checkout line. It also helps the FBI keep tabs on potential terrorist activity.

Biometrics - the ability to accurately identify people by unique traits and characteristics - has a tremendous range of public and private value, and the technology has been a critical tool to government agencies worldwide for quite some time. Thus, we're not surprised to hear that global demand is increasing.

According to Investor's Business Daily, the demand for biometrics hardware and software is expanding in light of high-level homeland security measures, border control and requirements for other government intelligence operations. The private sector is expanding as well. It looks like biometric technology is recession-proof: the International Biometric Group reports that the global biometrics market is expected to grow from $3.4 billion to $9.4 billion between 2009 and 2014.

Right now, fingerprint scanning is the biometric tool of choice for government agencies – the Department of Homeland Security's database alone holds 90 million sets of fingerprints. That said, global interest in face recognition is up; while fingerprint technology holds first place, face recognition is expected to be the second biggest revenue growth area in the biometrics industry in 2009, followed by iris scanning.

No More Smiling: Indiana Cracks Down on Happy Driver's License Photos

Latest news out of Indiana is that the state's Bureau of Motor Vehicles is instituting several new and somewhat interesting requirements for driver's license photos.

In an effort to aid facial recognition software and technology in its attempt to accurately identify residents, people will now be restricted from wearing scarves, hats and glasses, among other accessories.

Furthermore, drivers will need to refrain from smiling.

According to BMV spokesperson Dennis Rosebrough, it has been determined that smiles can inhibit the technology from accurate identifications and that the recognition software functions far better when given more standard images.

While this development may be a step in the right direction in terms of working to effectively leverage the technology at hand, a better approach would to to film a short video of the person, rather than take one single photo to rely upon for all identification purposes. While a single image is what always ends up on your license, there is actually no real reason that should be the only image the DMV can use when conducting a fraud search.

What we've found at 3VR is that comparing a collection of images of a person to other collection is always more accurate than working with a single passport-style photo-- regardless of facial expression, dress or accessories.

Multiple photos or even video would be a far more effective alternative, in our opinion, and what more, would even allow Hoosiers to resume their smiling at the DMV-- everybody wins.

Holiday Season Retail Fraud: Stores Going Proactive

Return fraud takes many forms, particularly during the holiday season: wardrobing (returning used goods as new or unworn, a trend that we reported on earlier this week), seeking refunds for stolen merchandise with fake receipts, even taking items off the shelf and carrying them up to the cashier to try to get store credit. This is all in addition to the ever-present act of shoplifting that skyrockets in times of desperation, synonymous with the period of lay-offs and recession we are currently experiencing.

An interesting aspect of this trend, however, is that although we've all heard about return fraud in the headlines, and might even know someone who has committed it, stores have been slow to react and only recently seem to be wising up to common retail fraud tactics.

It's about time.

According to the National Retail Federation, return fraud will cost retailers an estimated $11.8 billion in 2008 — $3.54 billion during the holiday season alone. Return fraud is expected to account for 7.5 percent of holiday returns – now that's a big chunk of change being ignored by retail institutions.

Forward-thinking vendors are now deploying both facial recognition technology and other advanced technologies to track and prevent return fraud and outright theft this holiday season. Databases have been developed to store photographs of known criminals and gang members that can be shared across retail institutions and linked directly into participating store camera networks. That means that when someone recognized from the database enters the store, cameras instantly alert security personnel. Furthermore, advanced search engines can be used to instantly sift through surveillance archives for any and all evidence related to a suspect.

During this holiday season, many institutions are taking this all one step further and even tracking and storing images of customers who seem to be taking excessive returns. Just a warning-- don't be unpleasantly surprised with a pair of silver bracelets on your next trip to the local shopping center.

Attention Shoppers: All Sales (Should Be) Final

While not the latest trend, the bear market has spurred a spike in acts of 'wardrobing' and other illegal retail fraud in stores nationwide.

'Wardrobing' essentially describes the act of purchasing a retail item, using the merchandise and returning it for a refund often with the original tags still attached. Think of it as ordering an entrée at a restaurant, having the waiter bring it to your table, then sending it back to the kitchen and having the waiter serve it to the next customer who orders the same meal.

As this blog notes, Good Morning America featured a clip last week on the year round practice that peaks during the holidays, as consumers are confronted by the traditional excess of events and cocktail parties and pressure to come up with something suitable to wear. The segment even has an interview with an anonymous 'wardrober' who admits she only committed her wrongdoings because she was strapped for cash.

While this woman and other frequent wardrobers often write off their misdeeds, it seems to be in their best interest to stop while they're ahead. Retail stores are finally catching on, expecting a loss of $11.8 billion in 2008 ($3.54 billion during the holiday season alone) due to fraudulent returns and thus increasing security surveillance. Furthermore, while some stores are extending return policies to garner more sales, many retailers are beginning to give bonuses to cashiers who catch wardrobers red-handed, as well as tracking those customers who are making a high number of returns.

So, wardrobers beware: if store cameras don't catch onto the illegal practice, few cashiers will pass up the bonus from denying a return of that stained shirt you wore to the company holiday party.

Why So Serious?

New biometric technology research out of Concordia University may prove useful in recognizing mood via facial features in public and highly-trafficked areas, such as airports and train stations. The system may also have implications on homeland security technology used to observe suspicious behaviors and characters.

Prabir Bhattacharya, a professor at Concordia's Institute for Information Systems Engineering, "has developed a computer image processing system capable of classifying human facial expressions and identifying what emotion a person is conveying."

"Whether certain points like the eyebrows or the lips are expanding or contracting, you can know what sort of emotion they are conveying," Bhattacharya said. In fact, his system is capable of identifying all ten expressions below. While 18 facial features exist to determine mood, his system only requires seven.


As the ITBusiness article states, "The system analyzes a facial expression by first measuring the distance between the eyes. Based on that, it is able to map out other regions of the face and set a template. Then, it can process different markers that give away a person's mood. By focusing on specific groups of muscles near the eyes, nose and mouth, the system determines mood without requiring a full facial profile. That means less data is needed to determine a profile than other types of facial recognition systems."


The big question remains -- what is the value of mood recognition technology in identifying suspicious individuals? While some critics maintain that surveillance is more focused on recognizing one individual versus another and continue to discount the value of identifying a person's mood, others point to its many applications in the security field. According to ITBusiness, the concept has already been utilized by Israeli security forces trained in psychological methods to evaluate mood in order to maintain order in large crowds of people.

Whether deployed at crowded train stations, endless airport security lines, or local banks across the country, mood recognition technology certainly has the potential to supplement current facial recognition technology and make surveillance systems all the more comprehensive.

However, it may be some time before this technology makes it out of the lab and into a video camera near you. Frankly, there just aren't many surveillance cameras that create the kinds of straight-on talking-head-style videos required for this kind of algorithm. Maybe we will first see mood detection in something like the self-check-in kiosks at airports so the airlines can know how we are feeling when we travel. Just remember to smile.

Face Rec Hacked! Needs "Liveness" Test

Very interesting article in CNet that highlighted an exciting new trend, but also pointed out that it may not be ready for prime time. Many new laptops, including new models from Lenovo, Asus and Toshiba, have started using facial recognition scans as the primary security mechanism for accessing their devices, rather than fingerprints or passwords. Definitely a cool use of new technologies, but as CNet points out, companies need to be sure they get it right before they introduce it to consumers, who would have no way to know their security was compromised.

In this test, security firm Vietnamese Internetwork Security Center (VISC) demonstrated vulnerabilities in laptops' face recognition-based authentication mechanisms that let anyone log in to a computer easily with a "special" photo of the legit owner, even at the highest authentication level. VISC was able to almost instantly produce a photo of CNet Editor Dong Ngo, taken over the laptop's webcam during a Skype chat, that fooled the computer's facial recognition software and successfully logged into a computer registered to Ngo.

Here's how Ngo described the offending photo:

About five minutes later, the technician produced a rather unflattering picture of me on a piece of letter-size paper. I could hardly agree that it was my mug on the photo. Nonetheless, when used in front of the laptop's camera, the Y430's authentication software was happy enough with the photo and logged in within a second. Pretty scary.

This type of hack is going to be very difficult for taditional facial recognition vendors to overcome. Early algorithms in this biometric field all focus exclusively on comparing one single image to another single image. Even if that image is being extracted from a laptop web camera. There is zero concept of context or "liveness" in this approach, and so it is easily spoofed. 3VR Security, is the only company I know of with a facial recognition platform built from the ground up to analyze streams of faces, like those in a video feed, rather than just single images. With this type of approach, subtle changes in motion, expression, pose, and other varialbles unique to a "live" 3D person can be analyzed at the same time a biomtric match is taking place and the kind of spoofing demonstrated here simply would not work. Maybe it's time for laptop vedors to upgrade their algorithms.

LPR Systems Abound Both in the US and Abroad

Whether to capture bank robbers and kidnappers on the run in the Cincinnati region or just to enforce local parking lot limitations in Aspen, License Plate Recognition (LPR) systems are spreading quickly throughout the country -- and as evidenced from recent headlines, they are finding their way overseas as well.

Yesterday, the Anti-Aggression Brigade from the Brussels Police Department in Belgium announced its deployment of AutoVu, Genetec’s IP LPR solution, to help detect stolen vehicles throughout the city. The day before, Hi-Tech Solutions Ltd. announced that its LPR solution, SeeCar, was selected to automate access control in 1,300 parking spots for AFCON’s Fast Park in the B.S.R. Towers in Israel.

While each initiative is meant to achieve a distinct objective — in Belgium, to track and find stolen vehicles, and in Israel to ensure that only permitted guests enter parking lots and to block unwelcome vehicles — these projects demonstrate the growing trend of LPR across major cities and ultimately the rise of urban surveillance worldwide.

Reported success has been limited for the numerous projects deployed thus far; however, recent innovation in camera technology, increased sensitivity to privacy infringements, and the hope of thwarting criminal attempts and increased protection of urban populations should indicate a promising future for LPR technologies.

Adobe Video-Object Manipulation Project Holds Significant Promise

Video Object Manipulation may be going mainstream.

Once the exclusive domain of computer vision geeks in the Department of Defense, and more recently Hollywood, Adobe has now got its hands on these algorithms and every YouTuber with an HD camera will soon have the tools to not only mark up video in fantastic new ways, but literally to bend reality.

What does it mean for security when just about anyone can add, remove or alter people and objects within a video stream to create a perfectly realistic video of something that never happened? Well, in my opinion it’s mostly not good.

But, there is a bright side -- the entrance of consumer-focused companies like Adobe into this industry is likely to help the security professionals as much as the criminals. More experts, new approaches, better tools and easier-to-use interfaces are a welcome addition to security’s video analytic offerings, and can certainly help security personnel fulfill the promise of improved surveillance.

CyberExtruder Gets 2D to 3D Face Patent

At the end of November, CyberExtruder announced that the company had been granted a new patent on their process for creating reliable 3D models of a person’s face from a single or series of 2D images.




Though the enforceability of this patent has yet to be tested, CyberExtruder’s 2D-to-3D conversion is certainly an important innovation to the security and biometrics technology industries. Why? It enables better matching between offline photographs and surveillance video. While the matching of "watch list" images against surveillance video has traditionally achieved mixed results as a result of inadequate lighting, angle, expression, etc., this patent could signify a leap forward in terms of the quality and value of 3D facial images.

In June of 2007, I wrote on how XID was using a similar technology in the “world’s largest” facial recognition access control project. In that instance, XID literally generated hundreds of thousands of variants of an enrolled employee’s face rather than using just a single 2D photo converted to a 3D model. Each day when an employee arrived for work, his or her photo would be taken and compared to the database of generated images rather than a single original. Interestingly, this approach generated huge improvements in the performance and accuracy of the Thailand access control system.



We’ll see if these two companies come into conflict over the new patent, but I don’t think they will. XID’s approach to 2D-to-3D is very different than CyberExtruder's-- almost quick and dirty by comparison. CyberExtruder, on the other hand, has become famous for its hyper-realistic…if sometimes creepy…generated floating 3D heads that lend themselves to applications well beyond security including gaming and movies, and even boast a fan in Phillip Rosedale of SecondLife.