h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

To answer this question...

>>Are there other pattern features that are all unique such as the pattern of convolutions on the cerbral cortex of the brain?

Well, yes...loads. Think about facial features, for example. They differ sufficiently from person to person to allow us to recognise individuals with reasonable accuracy. The problem is, though, that exactly how we manage this is still a bit of a mystery on it's difficult to automate reliably. Something like fingerprinting is more easily systemisable - there are some easily observable gross features.

Then there's iris recognition:
http://en.wikipedia.org/wiki/Iris_recognition

>>... Scottish police officer was wrongly convicted because her fingerprints...were judged to match those on the murder weapon. Similarly with DNA...<<
>>... the match was only on 3 or 4 points (if memory serves) and it is quite possible that completely different fingerprints may match on a few points. <<

Matching a few points is, of course, just fine - for an initial shortlist.
Presumably the point was then made that further matches should be sought, as a matter of routine, once that initial shortlist was identified?
What went wrong that that wasn't done (quite apart from that officer, if guilty, had left her prints in the full knowledge...)?

*3 or 4 points* from a large complex doesn't seem much to base a prosecution on (even to a layman like me).

>> Then there´s of course the koala bears. <<

But these are rarely left at the scene of a crime, too.

RF

apparently the full match is a 16 point match - for no scientifically justifiable reason! And her print was a smudged one so less points available to match anyway. The US seems to have a better system which is not points based but down to an accredited expert whose analysis must be backed up by two independant experts. They study the whole fingerprint not just selected markers.

Interestingly from that case the defence of how the Scottish experts got their result is that fingerprint evidence is a matter of opinion! Quite a claim given that most people consider it to be 100% - and this is the establishment saying this.

I´d imagine the matter of points is used when searching through databases. Once you then have it narrowed down, you can compare them to ech other properly.

That is certainly what the US crime drama series seem to show when they search the aphids. Although quite why greenfly should be experts in fingerprint recognition I ain't sure.

The difference is in the Scots case the 16pt match is *all* they use - and it was only through the introduction and review of a US expert (and sundry other foreign experts later) that she was found not guilty of perjury because those foreign experts used a different method which considered the whole fingerprint.

Thanks, Ictoan - that makes more sense.

Somehow, those 16 points sounds like someone's guesstimate & your statement
<< those foreign experts used a different method which considered the whole fingerprint. >>
sounds a tad more appropriate when considering criminal prosecution.