Skip to content or view screen version

The Register debunks Home Office statements on biometric reiliability

burntout | 19.10.2005 15:24 | Repression | Social Struggles

The Register has an article on the statistics behind multiple biometric tests.

From
 http://www.theregister.co.uk/2005/10/19/daugman_multi_biometrics/

"A regular correspondent (thanks, you know who you are) points us to some calculations by John Daugman, originator of the Daugman algorithms for iris recognition. These ought to provide disturbing reading for Home Office Ministers who casually claim that by using multiple biometrics you'll get a better result than by using just the one. Although that may seem logical, it turns out that it it isn't, necessarily."

And
""The key to resolving the apparent paradox," writes Daugman, "is that when two tests are combined, one of the resulting error rates (False Accept or False Reject rate) becomes better than that of the stronger of the two tests, while the other error rate becomes worse even than that of the weaker of the tests. If the two biometric tests differ significantly in their power, and each operates at its own cross-over point, then combining them gives significantly worse performance than relying solely on the stronger biometric.

This is of particular relevance to the Home Office's current case for use of multiple biometrics, because its argument is based on the use of three types of biometric, fingerprint, facial and iris, which are substantially different in power."

read the article for more skinny on it
here
 http://www.theregister.co.uk/2005/10/19/daugman_multi_biometrics/
and
 http://www.theregister.co.uk/2005/10/17/mcnulty_fingers_id_problem/

burntout

Comments

Hide the following 3 comments

politicians do not like clever people

19.10.2005 23:42

There is a nice analogy to this with algorithmic random-number generators. Generating high quality randomness mathematically is not trivial. Some bright sparks thought that it was obvious that one merely had to combine a bunch of known good generators, to give a really super-duper one. In reality, randomness is almost always massively degraded by such an approach.

In statistics, common sense is OFTEN wrong. Sadly, the scumbag politicians that use the services of mathematicians don't care about the truth, ONLY the lie that can successfully be sold to a badly educated public, for the direct benefit of said politicians.

twilight


Bad idea in principle

20.10.2005 19:38

The problem with ID cards and the national ID register isn't that there may be some technical problems with its implementation. The problem is that it is a fundamentally illiberal measure. People don't want their government to collect any more data on them than they already do, nor do they want to pay through the nose for them to attempt to do so.

Once you start getting into arguments about the implementation, you've lost. The details can always be sorted out. The technology will mature and no doubt come down in price. But it'll always be a dreadful idea.

Zorro


Yooo Hooooo, Twlight

21.10.2005 09:39

Twlight,

Zorro is spot on the mark. But listen, what puzzles me is your continuing failure to come up with the challenges put to you on other threads regarding your evidence. Just for ease of reference you may want to look at:

 http://www.indymedia.org.uk/en/2005/10/325872.html

 http://www.indymedia.org.uk/en/2005/10/325854.html

 http://www.indymedia.org.uk/en/2005/10/325677.html

A theory unsupported by evidence or proof is no more than a theory. Put up or shut up.

Observer