To: R. Jaynes who wrote (11638 ) 12/8/1998 10:47:00 AM From: David Read Replies (2) | Respond to of 26039
It looks like this CTST show may represent a real turning point for the biometric industry. Your link was very interesting, especially this excerpt: "'We break the problem down into two areas,' Morton said: 'private security, where you are only required to match a few locally stored prints; and public security, where potentially anyone's prints would need to be matched.' There is an immediate market for a personal-security system that would allow a gun only to be fired by its owner, or for a car-door-lock system that would only need to recognize family members." That describes a low-end application, such as Identicator is piloting at Fort Sill with biometric smart cards, or even lower, such as silicon chips. As an immediate matter, it challenges IDX F3 technology, which is CPU-based and high performance. These markets would allow someone to get the distributed recognition power needed without having to pay the premium for a high tech solution like F3. In the long run, in an integrated biometric world (where smart cards carry the user's ID between high and low use applications, and provide an authentication mechanism), there is a need for state of the art technology. But we don't know when that world arrives; it surely happens as the climax to a lot of other submarkets developing. The next excerpt from your link is wrong, in my opinion, since it implies the need for a one:many solution, and what is happening instead are one:one solutions with ID's provided by public key/private key, or computer sign-on options. "Public security poses problems since current methods require large central databases. If a customer makes a purchase with a credit card, his or her fingerprints would normally have to be matched against everyone who owns that particular card — unless there is a tamper-proof way of storing prints locally. Solving that problem would open up a range of applications." One tamper-proof way of storing prints locally is using a CPU-based peripheral, such as the F3. Another way is storing an encrypted template on the local PC, such as most competitors use, and IDT technology uses. Even F3 can store the template on a PC. It doesn't have to be all on a server. So the ultimate solution is a lot closer than this writer thinks. David PS -- I find it interesting that Microsoft and Sun are making technical presentations at CTST. Usually they say something worth listening to in these situations.