Underground Trolley Cameras Coming to Beantown, SweeperCams In Nation's Capital

Some interesting municipal surveillance initiatives over on the East Coast caught my eye recently:

First off, saw a Boston Globe article that the MBTA will begin testing cameras on the underground trolleys as part of a pilot program funded by DHS.

This is an ongoing effort that has already seen success. Three hundred buses and most of the T-subway stations are already equipped with cameras, and transit police have used video in almost 500 investigations, more than 240 of which have resulted in charges. This new program only will enhance the system and add more high-resolution cameras to improve video quality. (These cameras are facial recognition software-friendly and may be used to track people as well).

The video will be streamed in real-time to the Transit Police's operation control center in downtown Boston. According to the article:

"Our hope is that the cameras will be able to be viewed in a police cruiser, so that an officer responding to a call will have real-time viewing of what is happening on the scene," MacMillan said.

While station footage will be stored for almost a month, surveillance video from buses will only be kept for 72 hours. Not sure it's an issue with storage capacity due to high video quality, but a three-day limit severely restricts the video from being used in ongoing investigations, many of which last long past that time frame. However, this is a great effort on the part of the MBTA to take a proactive approach to increase transit security and identify repeat offenders.

Also interesting, I caught an article about D.C.'s Sweepercam implementation, a ticketing camera system installed on the city’s street sweepers to fine cars parked in designated-cleaning areas. The cameras are equipped with license plate recognition software so, should a car be parked in a designated spot, the cameras will note the plate number and the system will send the owners a ticket in the mail.

The system has already had a few hiccups and angered some folks with malfunctioning equipment and/or human error. Many operators have forgotten to turn off the cameras in places not slated for cleaning and thus, photographed legally-parked cars. In an email, DPW spokesperson Nancee Lyons responded to the mishaps:

“The camera may have captured the wrong vehicle, a duplicate ticket may have been issued, the camera may have been triggered accidentally and a ticket may have been issued during a day and hours when there was no street sweeping..."

With possible plans to extend parking meter hours to raise an estimated $9 million in extra funds for the transit system, I'm surprised San Francisco has not picked up on this one yet. DC has already dispensed 22,000 tickets from Sweepercam thus far -- that's good chunk of change that more than covers the investment.

Beverly Hills 48025

It seems that my old high school in Beverly Hills, Michigan has found itself embroiled in a debate over the use of surveillance in its hallways after cash, MP3 players and cell phones were stolen from several lockers. While this may signal a loss in trust in the community, it could also be reflective of the expensive nature of many items commonly found in lockers nowadays.

"It's one more sign that times are different today," Jim Ballard, executive director of the Michigan Association of Principals, said. "Years ago, we trusted each other more than we do now.”

I can also tell you firsthand that years ago, we students didn’t keep a lot of cash, MP3 players and cell phones in our lockers. Heck, if you broke into mine, you would have been lucky to find some old textbooks, gym socks and a Daisy Duke poster. Not quite a treasure trove of goods compared to what's readily available today.

But whether or not what students put into their lockers is valuable, the students themselves are pretty valuable, and it seems to me they deserve at least the same level of protection afforded to the malls they hang out in.

However, not all the students think new security measures are such a good idea.

Seaholm student Jake Drutchas wrote the following comment on the wall of a Facebook group called, "Seaholm and Groves Students Against Security Cameras," which has more than 870 members, "For now, let's start with a little student responsibility. Lock your stuff up. Don't leave it out in the open."

He has a point, but it’s also true that school security deployments have nearly always proven a smart development. I think Jake's argument is more founded on the fear that the same surveillance technology used to fight locker thefts will be turned on him for reasons other than safety.

Will the cameras be monitored all the time or just used to investigate incidents? Who will have access to the cameras and for what purposes? Lots and lots of very good questions.

If Beverly Hills Groves follows the national trend, they too will get security cameras installed -- if not this year, then sometime in the not-too-distant future. However, Jake Drutchas and his Facebook posse can continue to put their passion and energy on the issue to good use by helping the district draft a clear and responsible policy with respect to the use of surveillance inside their hallways.

There certainly are ways to balance security and privacy inside of educational institutions, and I think you will find the hippie baby-boomers who run school districts these days are more than open to them.

Opt-In Video Surveillance "Cloak" Aims to Calm Privacy Fears

The "Big Brother" argument -- that citizens are losing their individual privacy rights due to increased public security efforts -- is found in nearly all articles and reporting that cover the growing ubiquity of surveillance cameras worldwide.

Hewlett-Packard computer scientist, Jack Brassil, is hoping to address and resolve such criticisms with a new opt-in surveillance technology, Cloak, that he hopes will limit the privacy invasions presented by massive surveillance networks. According to Brassil, "Rather than prohibit surveillance, our system seeks to discourage surveillers distributing video without the authorization of the surveilled."

Cloak enables those who wish not to be identified in video footage to "opt-in" to the system, essentially making a "do-not-call" list for the surveillance grid. However, with this selection, the person must instead carry a 'privacy enabling device' that enables the image processing software to blur them out of any corresponding surveillance video, yet still allows the system to locate them on the grid.

Brassil is off to a great start in addressing the "loss of anonymity" argument that large-scale surveillance deployments bring to the forefront. However, the system still has a few kinks that need to be addressed before installation.

From a crime prevention standpoint, those that have the most to gain by blurring themselves out are those committing the crimes. In Brassil's proposal, it is impossible to 'un-blur' someone into focus for evidence in a possible investigation. Furthermore, requiring people to register GPS devices that tracks their location as a means to prevent surveillance systems from capturing their image seems to be just trading one modest privacy violation for another (perhaps even more) severe one.

As privacy analyst Ian Brown of the Oxford Internet Institute in the UK comments, "People shouldn't have to opt in to get privacy protection. And this system actively invades your privacy because it tells the service where you are at all times."

I commend Brassil in his efforts to resolve some of the privacy repercussions advanced surveillance deployments continue to introduce to society and have a couple suggestions to further improve the developing technology.

In an ideal system, every tracked face and motion activity should be blurred using a "reversible encryption" to enable future investigations if needed. That way, subject to policy (and perhaps subpoena), a person's anonymity can be reversed in a narrowly-focused way to solve crime and/or prove innocence.

Furthermore, the blurring functionality should be tailored to enable monitoring by police without revealing identity. For instance, security officials would be able to see if someone engages in violent or publicly-unacceptable behavior without identifying the individual. However, in an emergency, or if the system identifies a criminal, officials should have the option to disable the blurring feature in order to speed up the necessary response.

With these adjustments, Cloak is a much more workable proposal and more congruent with privacy, legal and societal concerns. Brassil is definitely a visionary in the space, and I hope to hear more about Cloak's progression in the near future.

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.

Video Evidence Doesn't Discriminate

Here's one for people who worry that surveillance video puts too much power in the hands of police. In this case, video helped a defendant (Guillermo Alarcon, Jr., pictured left) prove that police had planted evidence on him, or, at the very least had lied under oath about his arrest, which was enough to get charges against him dropped. We've all seen the movies, this is being painted as one of those cases where good cops go bad because the rules of their job keep them from doing a good enough job keeping bad guys off the street. Here, they planted cocaine on a gang member to make the charges stick. Some would argue that if you break the law to send a gang member to prison, the end justifies the means, but I think laws are laws and if your job is to enforce them you probably shouldn't be breaking them. Moreover, as an executive in the security surveillance industry, I think it would be just plain wrong if surveillance video were only used as evidence against certain wrongdoers and not others.

Why Fuzzy Footage Is Useless

The fact that even with several minutes of video footage the police are stilling calling a suspect in a recent Palm Springs robbery Hispanic or African American is a prime example of why poor quality surveillance systems are a huge waste of money. Even if they catch this guy, the video evidence will not be enough to prosecute him.

Here's hoping he cracks and confesses during the interogation.

Afghanistan To Get 'Big Brother' Cameras

The U.S. Government is contemplating a massive video surveillance project for the country of Afghanistan that would establish surveillance over all major thoroughfares in Kabul, the capital city, as well as all U.S. and multinational camps, traffic circles and Afghan ministry compounds.

The surveillance apparatus would provide a 24/7 command and control system that would be enable authorities to track personnel and identify vehicles with the use of license plate recognition systems.

This post from Wired does a good job of summing up my feelings. “Sorry, don't get me wrong; camera systems can be an extremely effective way of monitoring high-value or sensitive sites, particularly in Kabul. But a fiber optic ring around Afghanistan? License plate recognition systems? What the heck country are we talking about? How about a recognition system for donkey carts which are still commonly used in Afghanistan.”

Cam Case Collateral

The rapid proliferation of video cameras in the public, and their increasing importance in court room had brought forth a number of important ethical and legal questions for those “caught up in the net” of video surveillance.

Among the questions to be answered: can the cops resist the temptation to "go fishing" for other bad guys when reviewing tape of an incident? What are the ethical ramifications of overlooking minor infractions caught on tape and, most importantly, will the prospect of being filmed actually dissuade ne'er-do-wells from ne'er-do welling?

While we will never be able to completely eliminate these issues of “collateral damage” in surveillance-centric criminal investigations, we can do quite a bit with recent technology advances to minimize them. For instance, the days of needing to troll through days and days of video to track down a particular bad guy are over. Using searchable surveillance and biometric technologies, an investigator can target a video investigation on a particular individual. The video search engine returns only that video that relates to the subject in question.

Sometimes, modern surveillance technologies are viewed exclusively through the lens of ‘big brother,’ but in this case we can also see how those same technologies can be used to protect our privacy by bringing much needed focus and efficiency to video investigations.