Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Post: #1



From security point of view, fingerprints and biological data in
general constitute sensitive information that has to be protected.
Towards this direction, the method discussed in this paper isolates a
very small fraction of the userâ„¢s biological data, and only this
fraction is stored for future reference. This can also improve the
overall efficiency and bandwidth effectiveness of the system.
The novel application of computational geometry algorithms in the
fingerprint segmentation stage showed that the extracted feature
(characteristic polygon) may be used as a secure and accurate method
for fingerprint-based verification over the Internet. On the other hand
the proposed method promisingly allows very small false acceptance and
false rejection rates, as it is based on specific segmentation.
Biometrics technology allows determination and verification of ones
identity through physical characteristics. To put it simply, it turns
the human body in to his or her password. In this paper two algorithms
have been proposed by taking biometric techniques to authenticate an
ATM account holder, enabling a secure ATM by image processing.
Biometry, as the science of studying mathematical or statistical
properties in physiological and behavioural human characteristics, is
widely used in forensic and nonforensic applications in security field
such as remote computer access, access control to physical sites,
transaction authorization etc. In this paper the problem of fingerprint
verification via the Internet is investigated.
Specifically, the method that is used for the above purpose is based on
a traditional finger scanning technique, involving the analysis of
small unique marks of the finger image known as minutiae. Minutiae
points are the ridge endings or bifurcations branches of the finger
image. The relative position of these minutiae is used for comparison,
and according to empirical studies, two individuals will not have eight
or more common minutiae. A typical live-scan fingerprint will contain
30-40 minutiae. Other systems analyse tiny sweat pores on the finger
that, in the same way as minutiae, are uniquely positioned. Finger
scanning is not immune to environmental disturbance. As the image is
captured when the finger is touching the scanner device it is possible
that dirt, condition of the skin, pressure and alignment or rotation of
the finger all affect the quality of the fingerprint. Furthermore, such
methods may be subject to attacks by hackers when biometric features
are transferred via Internet.
In this paper a method is developed, which addresses the problem of the
rotation and alignment of the finger position. The proposed method is
based on computational geometry algorithms. The advantages of this
method are based on a novel processing method using specific extracted
features, which may be characterized as unique to each person. These
features depend exclusively on the pixels brightness degree for the
fingerprint image, in contrast to traditional methods where features
are extracted using techniques such as edge, minutiae points and ridges
detection. Specifically, these features express a specific geometric
area (convex layer) in which the dominant brightness value of the
fingerprint ranges. What makes biometrics useful for many applications
is that they can be stored in a database.
In brief, the proposed method is described in the following steps:
1. Pre-processing stage. The input image is made suitable for further
processing by image enhancement techniques using Matlab.
2. Processing stage. The data, which comes from step 1, is submitted to
specific segmentation (data sets) using computational geometry
algorithms implemented via Matlab. Thus, onion layers (convex polygons)
are created from these data sets, see figure 1.
3. Meta-processing stage (during registration only). The smallest layer
(convex polygon) of the constructed onion layers is isolated from the
fingerprint in vector form, see figure 2. For the rest of this paper,
this will be referred to as the referenced polygon. This is supposed to
be stored in a reference database, for subsequent verification.
4. Verification stage. This stage consists of the following steps:
1. An unknown fingerprint is submitted to the proposed processing
method (Steps 1 and 2), and a new set of onion layers is constructed.
2. The referenced polygon that has been extracted during registration
stage is intersected with the onion

layers and the system decides whether the tested vector identifies the
onion layers correctly or not.

Figure 1: Onion Layers of a set of points (coordinate vector).
2.1 Pre-processing stage
In this stage a fingerprint image, which is available from any of the
known image formats (tif, bmp, jpg, etc), is transformed into a matrix
(a two-dimensional array) of pixels. Consider, for example, the matrix
of pixel values of the aforementioned array. Then the brightness of
each point is proportional to the value of its pixel. This gives the
synthesized image of a bright square on a dark background. This value
is often derived from the output of an A/D converter. The matrix of
pixels, i.e. the fingerprint image, is usually square and an image will
be described as N x N m-bit pixels, where ËœNâ„¢ is the number of points
along the axes and Ëœmâ„¢ controls the number of brightness values. Using
m bits gives a range of 2m values, ranging from 0 to 2m“1. Thus, the
digital image may be denoted as the following compact matrix form:

The coordinate vector of the above matrix is:

Thus a vector S of dimension is constructed, which is then used in
the next
2.2 Processing stage
It is considered that the set of brightness values for each fingerprint
image contains a convex subset, which has a specific position in
relation to the original set. This position may be determined by using
a combination of computational geometry algorithms, which is known as
Onion Peeling Algorithms.
Consider the set of brightness values of a fingerprint image to be the
vector S (eq.2). The algorithm starts with a finite set of points in
the plane, and the following iterative process is considered. Let be
the set minus all the points on the boundary of the hull of S.
Similarly, define .
The process continues until the set is 3 (see figure 1). The hulls
are called the layers of the set, and the process of peeling
away the layers is called onion peeling for obvious reasons (see figure
1). Any point on is said to have onion depth, or just depth, i. Thus,
the points on the hull of the original set have depth 0 (see figure 1).
2.3 Meta-processing
In this case it is considered that the smallest convex layer that has
depth 3 (see figure 1) carries specific information, because this
position gives a geometrical interpretation of the average of the
fingerprint brightness. In other words, the smallest convex polygon
(layer) depicts a particular geometrical area in which this average
ranges. This feature may be characterized as unique to each fingerprint
because the two (2) following conditions are
1. The selected area layer is non-intersected with another layer.
2. The particular depth of the smallest layer is variable in each case.
Thus, from the proposed fingerprint processing method two (2) variables
are extracted: the area of the smallest onion layer and the depth of
this layer, which is a subset of the original fingerprint set S values.
Taking into account the specific features of the
aforementioned variables it is easy to ascertain that these may be used
for accurate fingerprint verification.
2.3 Verification stage
In this stage the subset is tested against a new subset set , which
came from the processing of another set N. This testing takes place at
the following 3 levels.
1. Subset is cross-correlated with subset .
2. The depths of the iterative procedure, from which the subsets were
extracted, are compared.
The intersection between subset convex layer and one of set S onion
layers is controlled.
Furthermore, it is considered that subset identifies set S as the
parent onion layers when:
1. The cross-correlation number of subset and subset is
approximately 1
2. The intersection between the convex layer of subset and one of the
onion layers of
set S is 0.
Otherwise, subset does not identify set S as the parent onion layers.

Figure 2: Theoretical presentation of the registration and verification
stages of two (2) onion layers.
Based on the feature extraction method and verification procedures
proposed in Section 2, it is described from a security point of view, a
model for a fingerprint verification system that takes place over the
Internet (see figure 3). There are two discrete stages for such a
system: a Registration Stage and a Verification Stage. Moreover, the
following components are employed:
Biometric Reader: it accepts a userâ„¢s analog fingerprint and transforms
it into digital information (e.g. TIFF format).
Processing Unit: takes as input the raw information provided by the
reader, and extracts the onion layers from the data. These are sent to
the Meta-processing Unit (during registration) or to the Comparison
Unit (during verification).
Meta-Processing Unit: it isolates the smallest convex polygon from any
set of onion layers it gets from the Processing Unit and submits the
referenced polygon to the Reference Database.
Comparison Unit: it intersects and compares the onion layers provided
by the Processing Unit with the referenced polygon provided by the
Reference Database.
Reference Database: it stores the usersâ„¢ reference polygons, provided
by the Meta-Processing Unit during registration, or provides the
Comparison Unit, during verification, with a userâ„¢s reference polygon.
All components must be tamper-resistant to avoid attacks by hackers who
wish to undermine the verification mechanism. Furthermore, in the
sequel we propose the use of some very basic cryptographic primitives
as well as several precautions in respect of securing communication
links between the units of the system.
All messages originated by all components of the system should be
digitally signed to avoid attacks such as man-in-the-middle attacks
that impersonate an entity to a component or vice versa. Such
impersonation (or spoofing) attacks are usually met in false acceptance

Figure 3. Communication Paths for a Biometric Verification System
Even the biometric reader should authenticate itself to the user, to
deal with ATM spoofing-like attacks, where a fake reader is used to
steal the userâ„¢s biological data. Furthermore, the digital signing of
data in conjunction with sufficient freshness information (timestamps,
serial numbers) can prevent various replay attacks. In such an attack
for example, the attacker feeds the component with digitally signed
data that he eavesdropped during a previous genuine verification.
Reasonably, encryption must also be used to protect the links between
units from eavesdropping or data injection.

Figure 4: Flowchart for Algorithm 1
4. Work on Image Processing:
Every system has its limitations. Therefore, identification based on
multiple biometrics is an emerging trend as Multimodel biometrics can
provide a more balanced solution to the security Multimodel systems
involve the use of more than one biometric system. For the contribution
to the above subject an algorithm is developed on banking security. For
this consider a bank using biometric technology for its security
purpose. The security is assured by using finger scan, voice scan, hand
geometry scan and by requesting the password given by the bank for a
particular user when necessary. The following are the flowcharts and
the algorithms.
4. Algorithm:
4.1 Algorithm 1:
a) STEP 1: A person enters the bank that uses biometric technology
(finger scan, voice scan & hand scan) for greater degree of security
(Figure 4).
b) STEP 2: The person is requested to give his
or her fingerprint (as input) on the finger scan pad.
c) STEP 3: The fingerprint from above step is compared with all
the fingerprints in the database.
If fingerprint is matched with any one of the fingerprints available in
the database (condition)
d) ELSE (i.e., if finger print does not find a match)
e) STEP 4: The person is requested to speak few words, which is
converted into digitalized code by the voice scanner.
f) STEP 5: The code in the above step is compared with all the voice
codes in the database
g) IF the code is matched (condition)
h) ELSE (i.e., if the code does not find a match)
i) STEP 6: The person is requested to place his hand above the hand
scanner so that the structure of the hand is recorded.
j) STEP 7: The data in the above step is compared with all the data
available in the database.
IF the data is matched
ELSE (i.e., the data does not find a match)¦¦.GOTO Algorithm 2.
k) STEP 8: access the matched file in the database
l) STEP 9: Exit.
4.2 Algorithm 2:

Figure 5: Flowchart for Algorithm 2
m) STEP 1: A request is sent to the database querying it to send the
password file from the OS security files (Figure 5).
The passwords are received in an encrypted form (DES-Data Encryption
n) STEP 2: The person is requested to speak his password.
o) STEP 3: The vocal password spoken in the above step is converted
into textual password by the speech processing circuit
p) STEP 4: This password is compared with the password file from STEP
q) IF match is found
THEN, ¦¦.. Access the database.
r) ELSE, ¦¦..Glow the danger light (indicating theft)
s) STEP 5: Exit.
Data stored in the Reference Database should also be encrypted and
protected against writing, to prevent a hacker from replacing a userâ„¢s
referenced polygon by his own in order to get false acceptance.
Taking into account the features described in Section 3 it is
ascertained that the method proposed in Section 2, having also in mind
the security considerations made in Section 4, can be used for accurate
and secure fingerprint verification purposes, because the proposed
feature extraction is based in a specific area in which the dominant
brightness value of the fingerprint ranges. On the other hand the
proposed method promisingly allows very small false acceptance and
false rejection rates, as it is based on specific segmentation. It has
to be noted that biometric applications will gain universal acceptance
in digital technologies only when the number of false rejections /
acceptances approach zero.
It has been pointed out that biometrics are not a security solution on
their own. For example, a well determined criminal could fake a
fingerprint using silicon imprints made from wax molds. However there
is an increasing trend to use biometrics in conjunction with other
technologies for security (pass codes or in attended environments). The
most promising application involves tamper-resistant smart-cards, where
the overall security is increased by unlocking a secret cryptographic
key only after a successful biometric verification.
Finally, more extensive experimentation is necessary, in order to
obtain statistically significant results and thus verify the conjecture
of this proposed method.
[1] A. K. Jain, A. Ross, & S. Pankanti, Fingerprint matching using
minutiae and texture features, Proc. International Conference on Image
Processing (ICIP), Thessalonica, GR, 2001, 282-285.
[2] D. Maio & Maltoni, Direct gray-scale minutiae detection in
fingerprints, IEEE Transactions on PAMI, 19(1), 1997, 27-40.
[3] L. Oâ„¢Gorman, Fingerprint verification, in
Biometrics (Jain, A, K. Bolle, R. & Pantanti, S.: Kluwer Acadenic
Publishers, 1999).
[4] T. Poon & P. Banerjee, Contemporary Optical Image Processing With
Matlab, (Hardcover: Elsevier Science Ltd, 2001).
[5] R. Bracewell, Two-Dimensional Imaging, (Horton M., NJ: Prentice “
Hall, Upper Sandle River, 1995).
[6] M. Nixon, A. Aguado, Feature Extraction and Image Processing,
(Butterworth Heineman, GB: Newnes-Oxford, 2002).
[7] R. Gonzales, R. Woods, Digital Image Processing,
[8] Anil K. Jain (Editor). Ruud Bolle, Sharath Pankanti. Biometrics,
Personal identification in Networked Society: Vol.479,

Important Note..!

If you are not satisfied with above reply ,..Please


So that we will collect data for you and will made reply to the request....OR try below "QUICK REPLY" box to add a reply to this page
Popular Searches: www nadacacheri verification com, passport verification based on passport, customer segmentation papad, segmentation ac, marget segmentation full mba project, image processing report, image segmentation ppt,

Quick Reply
Type your reply to this message here.

Image Verification
Image Verification
(case insensitive)
Please enter the text within the image on the left in to the text box below. This process is used to prevent automated posts.

Possibly Related Threads...
Thread: Author Replies: Views: Last Post
  imouse full report computer science technology 3 3,632 17-06-2016 12:16 PM
Last Post: ashwiniashok
  computer networks full report seminar topics 7 4,900 25-05-2016 02:07 PM
Last Post: dhanyavp
  Implementation of RSA Algorithm Using Client-Server full report seminar topics 6 4,991 10-05-2016 12:21 PM
Last Post: dhanyavp
  Optical Computer Full Seminar Report Download computer science crazy 43 34,277 29-04-2016 09:16 AM
Last Post: dhanyavp
  ethical hacking full report computer science technology 41 46,780 18-03-2016 04:51 PM
Last Post: seminar report asees
  broadband mobile full report project topics 7 2,102 27-02-2016 12:32 PM
Last Post: Prupleannuani
  steganography full report project report tiger 15 19,484 11-02-2016 02:02 PM
Last Post: seminar report asees
  Digital Signature Full Seminar Report Download computer science crazy 20 14,034 16-09-2015 02:51 PM
Last Post: seminar report asees
  Mobile Train Radio Communication ( Download Full Seminar Report ) computer science crazy 10 12,231 01-05-2015 03:36 PM
Last Post: seminar report asees
  service oriented architecture full report project report tiger 12 6,454 27-04-2015 01:48 PM
Last Post: seminar report asees