20-03-2012, 03:02 PM
Multi Purpose Code Generation Using Fingerprint Images
493 final.pdf (Size: 545.14 KB / Downloads: 70)
Introduction
Fingerprints today are considered as the most widely
used biometric features for personal identification.
Most automatic systems for fingerprint comparison are
based on minutia matching [3, 13]. The objectives of
biometric recognition are user convenience (e.g.,
money withdrawal without ATM card or PIN), better
security (e.g., difficult to forge access) and higher
efficiency (e.g., lower overhead for computer password
maintenance).
The Proposed System
The fingerprints image for any person can be utilized
to generate unique code. The fingerprint image passes
through a sequence of operations that includes
lightening, smoothing, edge detection and binarization,
then proceeds to thinning process and then an initial
code is generated, prior to final secure code
achievement. This section summarizes these processes
in few steps are shown below in algorithm 1. This
algorithm summarizes the proposed system steps to
produce the secure code. The whole process consists of
three types of actions; image preparation, feature
extraction and hashing. These actions are outlined in
the following sections in more details.
Image Preparations
Two major treatments are involved in fingerprint
preparation; image enhancement and thinning.
3.1. Image Enhancement
In order to normalize the starting image some
binarization-based approaches are applied. The
binarization and thinning processes must be preceded
by smoothing operation, based on convolution with a
mask [15, 16, 17]. The image enhancement phase
consists of lightening, smoothing, edge detection and
binarization processes. One filtering program written
in Delphi language is developed combining all these
operations.