On Wed, 17 Feb 1999 15:05:11 +0100 "Arjen G. Lentz" writes: >"Dr. Imre Bartfai" wrote: > >> only as an idea: >> I am also interested in fingerprint identifying. If I'd do it, I'd >store only >> the e. g. 32-bit CRC of the fingerprint to be identified. It hashes >pretty good >> (2^32 is about 4*10^9). > >Auch! No it doesn't. >Try fingerprinting with CRCs and you'll find a heck of a lot of >unknown people >with valid access ;-) You'll have much more of a problem of known people with slightly dirty fingers not getting access. Since changing one bit (or more than 1 but fewer than N bits, with N depending on the type of CRC) in the source data is *guaranteed* to chage the CRC, nearly every scan of the same finger will have a different CRC. It's not a viable way to "simplify" fingerprint images for storage. The comparison routine needs to be tolerant of small differences, but certain to reject large ones. CRC is just the opposite. I'd tend to agree with John Payson that the chance of different data having the same CRC is essentially random with the chance depending on the size of the CRC. CRC generators are like psuedo-random number generators, with the input data affecting the result. But CRCs tend to be rather small so the chance of a mis-match is high. True cryptographic "hash" functions are not so much designed for that purpose as they are believed to be non-reversible. Given the hashed output, it is extremely difficult to compute a fake input that will hash to that result. ___________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com/getjuno.html or call Juno at (800) 654-JUNO [654-5866]