South Korean Study Delivers a Face Rec First (90%+ Accuracy Using Surveillance Video)

Those following the South Korean government’s most recent biometric initiative know it to be one of the most ambitious facial recognition testing and deployment projects ever conceived. The project, in its various phases, has been featured quite extensively on Korean television here and here.

Under the project, the Korean National Police (NPA) sought to deploy facial recognition technology against Seoul’s plentiful video surveillance cameras, both to generate alerts to the presence of targeted individuals, as well as provide a mechanism to quickly search city surveillance archives for persons identified after the fact.

Though the NPA has been quite public on its facial recognition project in general, details regarding the specific technologies and testing results have been closely guarded. However, today after much anticipation, the NPA has finally made its initial findings public. According to the NPA and SK Networks spokesmen, 3VR's facial recognition platform demonstrated the highest percentage of accuracy of any of the eight tested technologies in trials run by the NPA and its testing partner, South Korea's IT giant SK Networks.

Indeed the results, at around 90% accuracy, would seem far better than those achieved in any previous public video facial recognition study. After a grueling multi-year testing process, in 3VR SmartRecorders and SmartCams provided between 85 percent and 92 percent accuracy in recognizing and matching faces in a few crowded, highly-trafficked public train stations in Seoul. In each case, the images analyzed were of fast-moving groups of commuters entering or exiting various transit areas en masse.

Said Sung-Ho Kong of SK Networks:

“In 2008, we performed a live, uncontrolled test of 3VR's facial recognition technology in Seoul subway stations, where the solution was an impressive 85-92 percent accurate, depending on conditions. No other solution approached this level of accuracy, vastly improving our ability to track, find and thwart crime in subways and other highly populated areas, which had previously proven extremely difficult to monitor.”

To better appreciate just how impressive the NPA’s results are, we need to compare them to the best previous facial recognition study. In 2006-2007, a similar facial recognition-use case was evaluated by the German Federal Police. After months of testing in a German subway with lighting and traffic much more controlled than in South Korea, a mediocre 60 percent accuracy rating was achieved.

The improvements in accuracy afforded by 3VR are the result of a unique approach to video facial recognition. While most face rec vendors offer technologies optimized for the comparison of flat, normally lit, passport-style photos, only 3VR's software was built from the ground up to address many of the unique challenges presented by real-time surveillance video.

According to Tim Frederick, director of engineering at 3VR:

“This study demonstrates the breakthrough power of 3VR’s patented facial surveillance technology. Unlike other attempts at high-volume face surveillance, which re-purposed still-image face recognition algorithms, the South Korean study benefited from 3VR’s end-to-end video analysis system, specifically designed for this type of demanding video application.”

Expect more details from 3VR and the NPA shortly, but in the meantime, please check out SDN's Rhianna Daniels' feature story on the announcement, as well as a quick diagram of how the technology works and even some up-close and personal screen grabs of the UI.

UPDATE June 22, 2009:

Here is a translation of the South Korean National Police's Phase 1 testing results:

Korean ATMs Exploring Face Rec Options

According to JoongAng Daily, one of three English-language daily newspapers in South Korea, the Korean National Police Agency is currently looking into new facial recognition technologies as part of a comprehensive public security plan to reduce crimes in the country.

Bank machines have been involved in several notable crimes in South Korea, including the highly-publicized murders by confessed serial killer Kang Ho-sun earlier this year. In those murders, an ATM was used to withdraw money from the victims' bank accounts before they were kidnapped and killed.

In the JoongAng Daily article discussing new attempts to further secure the nation's ATMs, the author discusses concerns around the implementation of facial recognition technologies, including problems with complicated instructions, cost and the technical headaches often involved:

"But ATMs with such advanced features still come with a litany of problems. Facial recognition machines were once test-run by several banks in 2005. Users were confused by a string of instructions such as "Raise your head, please" or "Step up to the machine, please," before being able to access their accounts. The disabled had even more trouble."

All of these arguments point to an interesting issue, and ultimately the resounding truth that facial recognition should not be implemented solely in the manner criticized above. Instead, government agencies should focus on using face finding technologies to help police procure better evidence, while also implementing face searching technologies in order to help track down and identify suspects.

Using these technologies, alerts can still be generated if someone with a mask or disguise attempts to use an ATM, but that notice would go directly to a monitoring center for evaluation, rather than immediately freezing all activity on the machine. As we know, when you adversely affect a customer's experience too much they become upset, and when too many people become upset the offending technology must be removed -- in this case, the removal wouldn't help anyone, and thus the complete freezing of ATM machines should be uniformly avoided.

3VR has created its own technology to address this problem by combining face finding, face matching and face searching. It is this synthesis of capabilities that ultimately makes the technology effective and leads to the best, least intrusive and most efficient results for all users -- from the smallest businesses looking to manage access control for a single door to multifaceted government agencies and airports.

Fool Me Once, Shame on You...

As reported last month, a South Korean woman fooled a fingerprint scanner at a Japanese airport to gain entry into the country. How did she do it? No need for an elaborate scheme -- instead, she used a special kind of tape purchased from a broker to easily bypass the state-of-the-art biometric system.

The 51-year-old woman used "special tape" to alter her fingerprints to trick the technology, which in November 2007 was installed at 30 airports around the country at a cost of around $45 million dollars. Designed to prevent entry by "international terrorists and other undesirables," reports say it is possible that a large number of South Koreans may have illegally entered the country using similar methods.

While the woman was eventually arrested in South Korea after attempting to purchase a fake passport, this incident raises a red flag about how effective border security really is and if expensive installments are worth the cost. In fact, last month, the DHS announced an update to its biometric technology in the US-VISIT program, which records biographic information to conduct security checks and verify the identities of international visitors to the United States.

However, why spend millions of tax dollars deploying systems that don't perform?

While biometric systems certainly complicate methods traditionally used by criminals to bypass the law, these tools will always exist -- ultimately, what customs and airport security checkpoints need is the deployment of an all-encompassing solution to stop those "special tape" users and repeat offenders in their tracks.

Integrating access control and facial recognition systems are two critical deployments needed to transform border security infrastructure and ensure a comprehensive solution to deter criminal acts. Designed to improve access control, passenger screening and liability protection, our deployments in airports (including Evansville Regional) monitor for criminal suspects and support both passenger and general airport visitor screening procedures with facial recognition technology, which stores profile pictures regardless of angle, shadow or lighting.

What does that mean in terms of catching criminals? We'll let's just say that no amount of "special tape" will get a person past an access control gate that requires an ID varified using facial recognition.

As the age-old proverb goes, "Fool me once, shame on you; fool me twice, shame on me." Multiple levels of security technology ensure the validity of this saying, and are ultimately crucial components of a successful surveillance program in the public places that need it most.