1. Fingerprint Recognition Using Correlation
Lalithkrishnan H
Gautham s
EEE department, R.M.K engineering college –kavaraipettai.
lalithkrishnanh@gmail.com
s_gautham1@rediffmail.com
Keywords: optical fingerprint identification, biometrics, artificial neural network, optical
correlation
characteristics. Physical biometrics is
Abstractn.:A major approach for based on the external characteristics of an
fingerprint recognition today is to individual, which is unique. Behavioral
extract minutia features from biometrics is based on the characteristics
fingerprints and to perform print of an individual at that instant. This may
matching based on minutia change over a period.
parirings. One of the most difficult However it has been known
problem in the fingerprint that there are number of people whose
recognition have been that the fingerprints could not be identified by the
recognition performance, which may feature based methods due to special skin
vary depending on environmental or condition, where feature points are hard to
personal causes, this paper discusses be extracted by image processing. Ratio of
the hybrid system based on optical people having this problem varies
preprocessor and artificial neural depending on sex, age, job grouping, etc.
network.
Addressing this problem, the
paper discusses the capabilities of hybrid
INTRODUCTION
system based on optical wavelet
Biometric identification
processing. Most digital processing
has been receiving extensive attention
methods for fingerprint recognition are
over the past decade with increasing
based on extraction of minutia features.
demands in automated personal
The advantage of this method is to
identification. Biometrics is to identify
identify based on the image processing
individuals using physical or behavioral
method instead of minutia features. The

2. major Minutia features of fingerprint
ridges are: ridge ending, bifurcation,
and short ridge (or dot). The ridge
ending is the point at which a ridge
terminates. Bifurcations are points at
which a single ridge splits into two
ridges. Short ridges (or dots) are ridges
which are significantly shorter than the
average ridge length on the fingerprint. fig 2 Bifurcation
Minutiae and patterns are very
important in the analysis of
CORRELATION
fingerprints since no two fingers have
been shown to be identical. Correlation is a
statistical technique that can show
The three basic
whether and how strongly pairs of
patterns of fingerprint ridges are the
variables are related. For example, height
arch, loop, and whorl. An arch is a
and weight are related; taller people tend
pattern where the ridges enter from one
to be heavier than shorter people are.
side of the finger, rise in the center
forming an arc, and then exit the other CALCULATING CORRELATION
side of the finger. The loop is a pattern Value of Interpretation
r (or rs)
where the ridges enter from one side of
a finger, form a curve, and tend to exit 1.0 Perfect correlation
from the same side they enter. In the 0 to 1 The two variables tend to
whorl pattern, ridges form circularly increase or decrease
together.
around a central point on the finger.
0.0 The two variables do not
vary together at all.
Now we're ready to compute the
correlation value. The formula for the
fig.1 Ridge ending

3. correlation is: input signal which is multiplied by some
filter.
In the Fourier
domain.whenever an image is incident
on the convex lens it is produce an exact
image having same fourier transform.It
stores its module using a computer and
this is compared with a sample.when
both the samples superimpose a planar
wave is created and recognition signal is
produced.
The optical processor
We use the symbol r to
receive the reflected beam of laser from
stand for the correlation.
the print and feds it to the rocesssing
CORRELATION COEFFICIENT system.The system to which it is fed can
The correlation be both within the computer or isolated
coefficient, r, ranges from -1 to +1. from the circuit. Normally for removing
The nonparametric Spearman the high frequency distortions.
correlation coefficient, abbreviated
rs, has the same range. Fingerprint Edges
enhancement
OPTICAL WAVELET
Wavelet ANN Module
PROCESSOR:
processing
An optical correlator
is a device for comparing two signals
by utilising the Fourier transforming
properties of a lens. It is commonly Fig 3.BLOCK DIAGRAM OF
used in optics for target tracking and WAVELET PROCESSING
identification.The correlator has an

4. 3.1)Low Pass Filtering: find the position where the correlation is
To get rid of maximal. This, aside from the training
the numerous high-frequency spikes period, is the most computationally
that seem to be present in the original expensive part of the entire algorithm.
images, we replace every pixel that The central region of the test image is
significantly deviates from the values then determined by selecting the central
of its four neighbors by the 65 x 65 patch corresponding to the
corresponding average. Filters do this. position of maximal correlation
3.2) Segmentation:. COMPRESSION AND
NORMALISATION
For each
image, we first draw a tight rectangular • Finally, each one of the two 65 x
box around each fingerprint using an 65 central regions is reduced to a
edge detection algorithm and 32 x 32 array by discrete
determine the geometric center of the convolution with a truncated
box. The central region of the gauss Ian of size 5 x 5.
reference image is then defined to be • This 32 x 32 compressed central
the 65 x 65 central square patch that region contains a low-resolution
occupies the region immediately below image, which corresponds
the previously described center. For roughly to 10 ridges in the
the test image, instead we select a original image.
similar but larger patch of size • The resulting pixel values are
105 x 105 (extending the previous conveniently normalized between
patch by 20 pixels in each direction). 0 and 1.
This larger patch is termed the • In our implementation, all the
window. parameters and in particular the
3.3) Alignment:. sizes of the various rectangular
We slide, pixel by boxes are adjustable.
pixel, the central region of the
reference image across the window of MAPPPING
the test image (by 20 pixels up, down
left and right) and compute at each step After the mapping is done the
the corresponding correlation, until we image is illuminated through light and

5. intensity of pixels at various points whether the image is a bright coloured
(x,y) are noted and scatter plot is image or light colored image. The gray
plotted and from this the linearity is shade distribution is noted in a histogram
studied.For example , Here F(xi ,yj) is called color histogram.
the pixel intensity or the gray scale
value at a point (xi ,yj) in the
undeformed image. G(xi* ,yj*) is the
gray scale value at a point (xi* ,yj*) in
the deformed image. and are mean
values of the intensity matrices F and
G, respectively. The coordinates or
grid points (xi ,yj) and (xi* ,yj*) are
related by the deformation that occurs
between the two images. If the motion
is perpendicular to the optical axis of fig 4. BEFORE CONVERSION
the camera, then the relation between (GRAY FORMAT)
* *
(xi ,yj) and (xi ,y )
j can be
approximated .
Here u and v are
translations of the center of the sub-
image in the X and Y directions,
respectively. The distances from the
center of the sub-image to the point (x,
y) are denoted by Δx and Δy. Thus, the
correlation coefficient rij is a function
of displacement components (u, v) and
displacement gradients . fig 5 AFTER CONVERSION (GRAY
FORMAT)
IMAGE CONVERSION
The major part in
this recognition is conversion of the
RGB format into gray shades. The
intensity of the gray shade denotes

6. and vertical axes to plot data points.
However, they have a very specific
purpose. Scatter plots show how much
one variable another affects. The
relationship between two variables is
called their correlation.
Scatter plots usually
consist of a large body of data. The
closer the data points come when plotted
to making a straight line, the higher the
correlation between the two variables, or
the stronger the relationship.
If the data points make a straight line
going from the origin out to high x- and
y-values, then the variables are said to
have a positive correlation. If the line
goes from a high-value on the y-axis
fig 6.HISTOGRAM PROCESSING down to a high-value on the x-axis, the
Onc variables have a negative correlation.
e the histogram is formed after METHODOLOGY OF
conversion, the scatter points are RECOGNITION
noted. To improve the accuracy of the
recognition it is better to select points Once the patterns are got by
towards the middle of the curve. These optical scanning, the above-mentioned
points are stored as template points. processing techniques are performed and
for particular displacement the intensity,
SCATTER PLOTS values are noted. These values are stored
as template. The same procedure is
Once the values are noted repeated for sample and for the same
for template and recognition purpose, displacement the intensity, values are
the values are plotted between noted. For both the intensity values the
respective pixels. These are similar to scatter plot is plotted. The correlation is
line graphs in that they use horizontal then done for the graph. If it is one then

7. sample matches .if it is –1 then the ADVANTAGES
sample does not match.
As a consequence
the following method is desirable:
• no contact with the specimen
required
• sufficient spatial resolution to
measure locally at the region of
interest
50
45 • the ability to capture non-uniform
40
intensity(templ
full-field deformations
35
ate)
30
• a direct measurement that does
25
not require recourse to a
20
15
numerical or analytical model.
10
5 REFERENCES:
0
1 2 3 4 5 intensity(sampl
i n t e n s i t y ( t e mp l a
te)
e) Ebooks
Fig 7. POSITIVE CORRELATION Correlation Pattern Recognition
BY: Kumar, B.V.K Vijaya.
60
(1)Biometrics for Network Security
50
40
By: Paul Reid ebook.
30 INTIENSITY(TEMPLA
TE) (2) “Biometrics recognition:
20 security and privacy concerns”
INTENSITY(SAMPLE)
10
S.Prabhakar, S. Pankanti,
0
1 2 3 4 5
I N TE N S I TY( TE M P L A TE ) IEEE security magazine.
(3)“High-speed fingerprint
Fig 8.NEGATIVE CORRELATION verification using an correlator”

8. A. Stoianov, C. Soutar