Page 1 of 1
An interesting statistic
Posted: Wed Nov 21, 2007 9:58 am
by Boyd Baumgartner
From an article I read today.
RE: the US VISIT program
Travellers to the United States, meanwhile, will soon be required to submit to a test of all 10 fingerprints. US authorities have for the past four years taken prints of the index fingers of visitors to the US, creating a database of 90 million individuals. However, the limited fingerprinting generated up to 70 false matches a day, delaying travellers while they proved their identity, officials said yesterday.
I'd be interested in what they mean by 'false match'.
The article then goes on to say:
The director of the department's US-VISIT program, Robert Mocny, said 2000 individuals had been stopped from entering the US because their index fingerprints had not matched those recorded for their passports.
"By taking all 10 fingerprints, that makes that person unique," Mr Mocny said. "There will be much less sending back to secondary [identification], much less false matches."
I like the second paragraph that implies that the current method of taking two finger impressions fails to meet the threshold of uniqueness, but all ten fingers,
that passes the bar.
It would seem that this is a poorly worded article and the author is using the word 'false match' to mean 'failed to match'.
Does anyone know anything about the program? Is it a lights out operation or does 'false match' just mean 'requires human intervention'.
However claiming 70 'false matches' a day seems to be a dangerous choices of words these days.
http://www.smh.com.au/news/world/finger ... 67148.html
Posted: Wed Nov 21, 2007 10:34 am
by Pat A. Wertheim
Several things here strike me --
First, the article is quoted as saying that a false match resulted in a delay while the traveller proved his/her identity. That tells me it was not a positive identification or individualization, merely that the computer (AFIS?) generated a high enough score to require the airline security personnel to check a traveller's identification more closely.
Second, I strongly suspect there are no qualified fingerprint experts on duty working this system at the airport security stations, only the usual security personnel who check bags, work the metal detectors, run the explosives and narcotics trace detection devices, etc.
Third, I would think that, if questioned in court about this article, it would be easy enough to explain that attacking a fingerprint individualization by a qualified latent print examiner by comparing what we do to this airport screening system is truly an apples to oranges comparison. The terminology is different, the methodology is different, the training and experience of the people working the device is different, and the intended use of the system is different. There is no "match" here.
Posted: Thu Nov 22, 2007 4:03 am
by Daktari
Interestingly German shoppers appear to be ditching plastic in favour of fingerprints. To see more follow the link. (It’s in English!)
http://news.bbc.co.uk/player/nol/newsid ... id=7102539
Wonder if a certain wee shop in Troon will follow suit?
Posted: Fri Nov 23, 2007 2:44 pm
by Justice Pie
Found this article interesting and looked a little into Munro...
Ian Munro - New York
October 25, 2007 - 12:53PM
Ian Munro
Photo: John Woudstra
Ian Munro is the New York correspondent for The Age and The Sydney Morning Herald.
Previously, he was The Age's law and justice editor and senior writer on major crime stories, legal issues and national security.
He has been a journalist since 1980, having worked with Leader Associated Newspapers, the Herald and Weekly Times and The Age.
During his career he has covered state politics, urban affairs, the law and business as a news, general and feature writer.
His positions have included state political reporter for the Sun News-Pictorial, urban affairs editor and senior writer for The Age.
He is a former news editor of The Sunday Age, and was for more than three years principal press secretary to two premiers of Victoria, John Cain and Joan Kirner.
He is, with Age colleague Gary Tippet, the author of a collection of true crime journalism.
http://www.theage.com.au/articles/2007/ ... 09706.html
And not sure if this is the same Ian Munro...
AUSTRALIA: Warrant served on Age journalist
Police obtain search warrant to reprosecute Jack Thomas for alleged terrorism
The Age
Monday, September 4, 2006
By Vanessa Burrow
Federal police have served a search warrant on an Age journalist in an attempt to reprosecute Jack Thomas for alleged terrorist activities.
They also have a warrant to seize documents and tapes from the ABC's Four Corners program.
Age reporter Ian Munro handed over 20 pages of shorthand notes taken when he interviewed Thomas on several occasions while he was on bail this year.
The notes were the basis for a report titled "Islam's Warrior," which appeared in The Age on February 27, the day Thomas was convicted of receiving funds from al-Qaeda and holding a false passport.
The Victorian Court of Appeal overturned the conviction on August 18. The evidence used to convict Thomas, collected in an interview in Pakistan with Australian Federal Police, was ruled inadmissible.
The Director of Public Prosecutions has requested a retrial and, in an application to the court, lawyers cited evidence from an interview Thomas gave to Four Corners.
Its report, titled "The Convert," was compiled by journalist Sally Neighbour and screened on February 27, the same day Munro's article appeared in The Age.
In the article, Thomas admitted to having met al-Qaeda organiser Khaled bin Attash and being provided with money and an air ticket home.
He also said a Taliban-issued visa to Afghanistan had been doctored in his passport to look like a Pakistan visa.
Yesterday, an Australian Federal Police spokeswoman said journalists were not regularly asked to hand over their notes.
"It doesn't happen that often, but if there's material that we need it's the same process for members of the public as it is for media organisations," she said.
The warrant ordered Munro to surrender any information related to Thomas and several organisations, including Pakistan's Inter-Services Intelligence Directorate, the Guantanamo Military Facility and the Federal Police. It said there were reasonable grounds for suspecting the information would "afford evidence" that Thomas received funds from a terrorist organisation and held an Australian passport that had been falsified.
Date Posted: 9/4/2006
http://www.asiamedia.ucla.edu/article-p ... ntid=52167
Article in question...
Islam's Warrior
entering the US
Posted: Fri Nov 30, 2007 3:03 am
by karen M
the kind of non matches will probably encompass what happened to my friends father....
he had a lovely holiday in the US about 3 years ago, had his finger scanned, and went away home to Scotland.
He then, had an accident in his garden shed involving a circular saw, which conveniently sliced off the same finger.
Early this year, he decided to have another lovely holiday in the US.
Took him over 6 hours to get into the country, as the security staff just did not know what to do.
He still enjoyed his holiday, and can laugh about it now.
Karen
Posted: Mon Dec 03, 2007 5:37 am
by Ann Horsman
I think the Customs live scan only has the capacity to scan and search index fingers which might have caused the probelm for your friend's father. I went in to the local US Customs office not long ago to get a Form 4457 to take a firearm to Canada for a hunting trip.
We need the form to prove ownership upon returning to US soil. While there I observed another Customs Agent collecting fingerprints with their issued equipment. It appeared to me then that only the index finger is captured.
I can wholeheartedly say I was disappointed and concerned to see how poor the equipment was and how limited it is. It's a shame that the American taxpayer isn't getting their money's worth.
I have heard the Feds may be switching to a 10 print system, let's hope so.
Posted: Mon Dec 03, 2007 5:39 pm
by mdavis
Fingerprint scan devices used for security purposes, whether it's to activate your laptop or enter a facility or country, are very simple compared to most AFIS systems in use today. They depend on comparison of one known finger against a small database for confirmation of "likeness." This is worlds different than comparing a latent print against a potentially huge database of all possible offenders. The chances of a single, random, unauthorized person having a statistically similar fingerprint to a small stored sample database is quite low. Bioscan systems also do not have to deal with marginal quality unintentionally deposited latents or distortion caused by non-optimal surfaces.
Posted: Mon Dec 03, 2007 6:10 pm
by Pat A. Wertheim
"Morning Edition" on National Public Radio did a story this morning in which they said that iris scans are more accurate than fingerprints. While I didn't really like that statement, maybe in the case of the single-finger systems used by Homeland Security at airports, they might be correct.
Posted: Tue Dec 04, 2007 5:16 pm
by mdavis
Fantastic idea! If we can figure a way to have criminals leave an iris impression, maybe by peering through the peephole to check for occupancy, we can eliminate the less accurate latent print community.
It would be interesting to see the algorithm used by iris scanning software.
The AFIS systems with which I am familiar reduce an analog image of a fingerprint to a digital representation of L2Ds. They do this by attempting to cut out "bad" areas (smeared, pattern edges, etc.), blurring the analog image into a high contrast black and white line drawing, digitize the lines to a "skeleton", mark the L2Ds as X,Y,theta coordinates in algebraic quadrants and store the result. Essentially all L3D is lost and/or ignored. In addition, search algorithms build in a "slop" factor to attempt to account for skin flexibility and pressure distortion thus widening the number of potential "hit" scores, but at the same time, lessening the accuracy of the search. I would expect an iris program to be similar, but without the possibility of much, if any, distortion, which would theoretically increase the precision with which the search and "comparison' functions are able to work.