Technologies Help Protect Mass Transit against Terrorist Attack
Some ways to 'harden' the
vulnerabilities of mass transit operations
So called 'security' cameras can be found in growing
numbers in many major cities. New technologies can
automatically monitor the video feeds and may detect both
known terrorists and anyone else deemed to be acting
The huge numbers of daily
commuters and extensive public transport networks make it close
to impossible for traditional technologies and security
procedures to detect terrorists and prevent suicide bombings.
Some new technologies promise
to offer some automated help that may help detect terrorists,
although whether such detection will be sufficiently timely as
to assist in their apprehension prior to committing their act of
terror is uncertain.
For the best approach to
securing our public transport system, please read on to the
final part of this series.
This is part four of a
five part series on the risks in mass transit systems and how to
protect against them. If you've directly landed on this
page from a search engine, you might wish to start at
the beginning of the series and
The previous part of this
series showed how traditional
methods of so-called 'security' are of little or no use when
attempting to protect mass transit systems against suicide
Happily, there are at least four new
technologies at various stages of development and deployment
that could help secure our mass transit systems.
Automatic computerized video monitoring
New programs are being
developed whereby computers will automatically detect suspicious
behavior via video feeds.
These programs can detect
obvious threat activities such as a person leaving a bag
somewhere and then going away without the bag. They also
claim to be getting better at recognizing giveaway traits on the
part of passengers that indicate nervousness or evil intent.
There are problems with
these types of technologies. First, with crowds of people,
the ability of cameras to track individuals is limited, because
they may be obscured from the camera view by other people,
making it harder for an accumulation of data.
Secondly, apart from a wait
on a platform, a person is moving through a station, making it
harder for a computer monitoring program to follow each
individual from camera to camera, and for giveaway behavior to
manifest itself. People are more likely to act
suspiciously when loitering/waiting than when focused on the
task of walking in a group of people from somewhere to somewhere
Thirdly, if the systems are
tweaked to make them reasonably sensitive to suspicious
behaviors, there is likely to be a high incidence of 'false
positives' where they incorrectly guess that innocent passengers
may be exhibiting threatening behaviors. This not only
inconveniences/annoys the passengers, but drains personnel
resources from continuing to seek out real bona fide threats.
It also means that the staff
sent to investigate possibly suspicious people lose their edge.
After investigating 500 false alarms in a row, how alert will
they be when approaching the 501st alarm?
Fourthly, these systems tend to
instill an aura of misplaced confidence in the security people,
such that they relax their guard and are not as vigilant in
other parts of the detection process.
Fifthly, there's the lead
time issue - by the time the computer has sounded an alarm, and
by the time a real person has responded to that alarm and
directed security personnel to the person of interest, it might
be too late.
Lastly, and just like with
behavior detection done by humans, the purposeful and brief
transit through a subway station gives fewer opportunities for
meaningful behavior patterns to manifest.
This technique has been
employed in Las Vegas casinos for some years now - computers can
automatically recognize 'people of interest' when sighted
through video feeds.
But this recognition process
presumes that the terrorists in question are known terrorists
and that good facial imagery has been fed into the monitoring
computer so they can then be recognized with some degree of
accuracy. These two requirements are not things that are always
present in suitably optimized form.
Even when the terrorists are
known to be bad guys and their images are loaded into a facial
recognition system, bad light, obscured facial images by other
people in the crowd, and some types of facial disguise may all
still allow a terrorist to slip by.
This type of system is of
vulnerable to lead time issues.
A type of remote invisible
explosive scanner is being trialed in the UK. It uses a UV
light source to scan people, taking a couple of seconds to
detect the possible presence of any explosive residue.
It will probably be possible
to do this, to some hopefully sufficient extent, as people pass
through fare turnstiles.
Unfortunately, UV light can
damage a person's eyesight, and also harm their skin and
accelerate the possibility of skin cancer. It is not known
what intensity of UV light such new systems may have or what the
accumulative effect may be on commuters being exposed to it
several times a day.
This also requires the
terrorists to have conveniently contaminated themselves with the explosives
they are carrying.
Trusted Passenger programs
To flip things around, it
may be possible to take passenger tracking systems already in
place, such as London's Oyster card, and to enhance these to
allow for designated trusted passengers.
Anything to reduce the
number of unknowns is clearly a good thing, but a trusted
passenger program would only apply when going into subway
stations. Once through the turnstiles, all passengers merge into
one big mass of passengers.
tracking/evaluation systems could know to ignore trusted
passengers and focus more carefully on unknown passengers, and
unknown passengers could receive more careful scrutiny when
The cost of establishing and
running a trusted passenger system would be great, and the
returns of uncertain value, and the privacy tradeoffs - the
transit authority would then get to know of all your travels -
It may be a good
enhancement, and may offer some cumulative additional security,
but as a freestanding system with nothing else, it is next to
So, with only one part of
this article series, what should be done? At last, the
answer to that question is finally only
one more click away.
If so, please donate to keep the website free and fund the addition of more articles like this. Any help is most appreciated - simply click below to securely send a contribution through a credit card and Paypal.
2 Apr 2010, last update
02 Jul 2017
You may freely reproduce or distribute this article for noncommercial purposes as long as you give credit to me as original writer.