Brad,
"If I understand you correctly......the person who is issued a card....puts down a fingerprint......and the print IS entered on the card....for future possible use?"
Yes
Don't they need a data base to compare the fingerprint to?.....or a way to save the print for future reference?"
Well, that depends on the architecture. For instance, they could send the reference template off to a central data bank, under the label "Private Joseph X. Smith." In that case, when Pfc. Smith shows up, he puts his finger down on a scanner, identifies himself, and the system looks up the template under that label and sees if the present scan matches the one on permanent record. However, I think it may be more likely that Pfc. Smith would store his reference template on the smart card itself. Later, he would put the smart card into a reader while putting his finger down on a scanner. The system would then see if there were a match (by doing the matching off-card at some distant location) and, if so, would unlock the card. Neither version depends on AFIS technology, and each requires only one finger checked against a labelled reference template.
I got the idea that these cards must be ready for future biometric uses.....doesn't this require the card maker to "build in" biometric functions? I would think it would.
The smart card itself doesn't require onboard extraction or matching algorithms, since it will get the power for those functions supplied by a connected PC when the card is inserted in a PC-connected card reader that also has a DFR- capability via a peripheral, keyboard, or mouse. But I would expect the card to have some dedicated, coordinated memory corner to store biometric information; that way another part of the system could call for the biometric information specifically.
I thought that's what our deals with Infinion and Motorola were all about. Are there no royalties for IDX if the IDX bio-engine is used in these cards?
The Infineon and MOT deals cover different areas. Infineon, as the maker of a silicon chip, hopes to be able to create a robust, inexpensive reader that can go on a smart card. So does Veridicom, for that matter. (IDX optical technology is not going to get skinny enough to get placed on a credit card sized token.) A reader on a smart card would do away with the need for a PIN or an associated reader on a PC just to unlock the contents of the smart card, if the card could support extraction and matching algorithms. That is far in the future, if ever. The MOT deal has to do with putting readers in everything short of a smart card: desktops, laptops, cell phones, PDAs. No matter what version you use -- local card or remote Internet -- the extraction and matching algorithms and protocols are what BioEngine provides.
So is the only revenue chance here in the hardware?
No. There should be minimum revenues at enrollment stage here. It's the implementation stage where the money comes in. Suppose IDX lands a 4 million seat deal with the DoD @, say, $10 per seat per year for use of itrust services. That's $40M in annually recurring revenues with an extremely high margin.
"And doesn't the Army have a data base of fingerprints of their soldiers?......I would think so. Why would they not run a search of their master data base at the time of card issuance? Doesn't the Army run one to many searches?"
I don't know if the Army tenprints recruits. If they do, it's an AFIS system. However, that's not a system you would use for authentication/verification purposes. It's a system you would use if you think the recruit may be hiding his identity or his background and you need to check against a 1:many database.
________________
OT: I read somewhere that one Florida newspaper recount has Bush with a statewide lead of one vote. (I think it was Justice Scalia again.) |
