IBM Patenting Airport Profiling Technology 129
An anonymous reader writes "InformationWeek's Wolfe's Den reports that IBM has filed a dozen applications to patent a sophisticated airport security system which supports passive software-based profiling of potentially dangerous passengers off of pre-programmed rules. The setup uses a collection of sensors — video, motion, biometric and even olfactory — in terminals and around the airport perimeter, to supply raw data. 'These patents are built on the inference engine, which [analyzes sensor data and] has the ability to calculate very large data sets in real time,' says co-inventor Roger Angell. A small grid of networked computers delivers the necessary processing power. Two applications go one better than Israeli-style security, analyzing furtive glances to detect, according to the title of the patent application, 'Behavioral Deviations by Measuring Eye Movements,' as well as measuring respiratory patterns."
Patents (Score:1, Interesting)
The new whipping boy that SCO has fallen off the narrow /. radar. Hoo-freaking-ray!!!
This sucks (Score:2, Interesting)
The need for this just appalls me. Hate it. It's amazing what a small group of "dedicated" people can do with a few airplanes.
I feel horribly for the loss of life, but I can't imagine those terrorists ever expected it to get this far.
Stupid.
This is all marketing hype and the patent would... (Score:4, Interesting)
...(if granted) never stand up in court unless something truly novel was listed because this sort of 'data fusion' has been going on in the security industry for the past 10 years.
There is a very specific reason you will only see this sort of 'product' in testing for the next 10 years - 'false positives.' That's a very very important phrase in the security industry because software based solutions are supposed to act as force multipliers (although historically they're used to reduce forces in order to lower costs through automation, not to augment it) and if you've a high 'false positive' rate (as ALL of these behavioral analysis systems do) you actually impede normal security operations. Research in this area of physical security is active and ongoing, but veyr unlikely to produce anything usable except in very specific scenarios (objects left behind, loitering, et cetera.)
Re:Money, Money, Money (Score:2, Interesting)
Also, I believe this is meant to be used to ASSIST actual people at checkpoints. It can't hurt to have an additional system help pick out suspicious people. Humans can only look at one person at a time as they walk by but the machine can keep a close eye on EVERYONE. What about people that try to get by the humans at the checkpoint and are worked up and nervous but then take a deep breath and regain their composure when they get close to the TSA employees? With a computer they'll be able to monitor and pick out those people well before they even get to the Xray machine and mark them for screening.
Sounds like it can't hurt. It's not like if the machine gives a false positive they throw you in jail. They just use a wand or pat you down. Heck, I was patted down the last two times I went through security and it barely took a minute.
Comment removed (Score:2, Interesting)
Re:This sucks (Score:5, Interesting)
Israel has been dealing for this threat on a much higher level for years. It's not as hassle free a solution as no security, but the wait times are substantially less, and success substantially better than America's Funniest Security Theater.
Thanks to the
Israelification of American Airports [thestar.com]
Re:This is all marketing hype and the patent would (Score:3, Interesting)
However, when we used our data fusion algorithms to augment the history of a person at a checkpoint (simulated) false positives were okay, they just enhanced the interrogation.
The problem is false negatives which is much harder to quantify. Of course we had access to a scintillator that could identify trace radioactive potassium from the banana you had for lunch...