25-07-2012, 01:07 PM
Personal Authentication Using 3-D Finger Geometry
3d finger.pdf (Size: 842.06 KB / Downloads: 31)
INTRODUCTION
BIOMETRIC authentication, once used for granting access
to high security infrastructures, is gradually finding place
in a wider range of applications. However, until today the requirement
for highly reliable authentication has led to compromises
with respect to user acceptance. It is clear that reliability
and user convenience should coexist in order to achieve a widespread
acceptance of biometrics.
The work in this paper is partly motivated by applications
where the convenience of the user is the first priority. These
applications include personalization of services (home, office,
car) and attendance tracking in working environments. A user
authentication system based on measurements of three-dimensional
(3-D) hand geometry is proposed. Unlike other hand geometry
verification techniques the proposed system is less obtrusive.
PREVIOUS WORK
Hand geometry recognition is one of the most popular biometrics
used today for user verification. It works by comparing the
3-D geometry of the hand with a previously enrolled sample. A
simple two-dimensional (2-D) camera sensor is commonly used
to capture an image of the user’s palm, while a lateral viewof the
hand is captured on the same CCD thanks to a mirror. The user
has to put his/her hand on a special platter with knobs or pegs that
constrain the placing of the hand on the platter. This greatly simplifies
the process of feature extraction performed by analyzing
the image contours of the hand views [1], [2]. Various features
such as width of the fingers, length of the fingers and width of the
palm have been proposed. Satisfactory recognition results are obtained
(96% for recognition and less than 5% EER 1 for authentication).
HAND DETECTION
The first step is the segmentation of the hand from the body,
which is achieved using available depth information and exploiting
a priori knowledge of the human body geometric structure
and the authentication scenario.
The distance of the user’s hand from his/her face is not guaranteed
tobesufficiently large(usuallybetween5–20cm).Therefore
we may not rely on simple thresholding to separate the hand from
the face. Detection and segmentation of the hand is based on a
more elaborate scheme that relies on statistical modeling of the
hand, arm and head plus torso points in 3-D space.
FEATURE EXTRACTION AND MATCHING
We use the above signature functions to extract 3-D geometric
features of the fingers and we utilize these features for similarity
measurement. Altogether, ninety six measurements quasiinvariant
to hand position, orientation and finger bending are
used. The thumb is excluded since its measurements are unreliable.
Twelve width and twelve curvature measurements are
computed for each of the remaining four fingers. The length
of each finger is estimated by the largest computed distance
from the corresponding finger-tip.