By Steve Ranger, 20 July 2006 13:25
NEWS
Accuracy
Some biometrics are more accurate than others. Iris is generally considered to be the most accurate, followed by fingerprint then face.
Accuracy is usually determined by measuring the False Acceptance Rate - i.e. allowing in imposters - and the False Rejection Rate, which is when an authorised user is barred.
The list from A to Z
Click on the links below to find out more on each of the entries in the list.
A is for Accuracy
B is for Behavioural biometric
C is for Cash machine
D is for Database
E is for Ear
F is for Facial recognition
G is for Gummi bears
H is for Hand geometry
I is for Iris
J is for Juan Vucetich
K is for Keystroke dynamics
L is for Liveness testing
M is for Mobile phones
N is for Network security
O is for Oxford
P is for Palm
Q is for Queues
R is for Registration
S is for Signature verification
T is for Twins
U is for Universality
V is for Voice verification
W is for Walk
X is for X-ray
Y is for Young
Z is for Zurich Airport
Accuracy will depend on the particular technology or mix of technologies involved and also the conditions under which they are being measured. For example, it's easier to capture a voice pattern in a quiet room than on a noisy street.
Comments
There are 3 comments. Join the discussion
1. Karen Challinor
It would be interesting to know which of these have published positive and negative failure rates
2. Mike Murray
Biometric Access Protection Devices and their Programs Put to the Test
http://www.heise.de/ct/english/02/11/114/
3. Martin George
Iris recognition has actually been around for more than 20 years - a key patent "Flom and Safir" expired in Feb 2006. This has opened to door to commercial competition. Companies like ours are opening up the market with low cost camera devices, and low cost business models for algorithm deployment on devices and on server farms, where large scale cross-matching must take place for ID applications like Cards and Border Control.