IOSR Journal of Electronics and Communication Engineering (IOSRJECE)

ISSN : 2278-2834 Volume 2, Issue 6 (Sep-Oct 2012), PP 17-21

www.iosrjournals.org

Fingerprint Recognition Using Gabor Filter And Frequency

Domain Filtering
Kondreddi Gopi1, J.T Pramod2
1,2

(ECE Department, MITS, Madanapalle, India)

ABSTRACT: Fingerprint recognition is one of the most popular methods used for identification with greater
degree of success. The fingerprint has unique characteristics called minutiae, which are points where a curve
track finishes, intersect or branches off. In this work a novel method for Fingerprint recognition is considered
using a combination of Fast Fourier Transform (FFT) and Gabor Filters to enhancement the image. The
proposed method involves combination of Gabor filter and Frequency domain filtering for enhancing the
fingerprint. With eight different orientations of Gabor filter, features of the fingerprint are extracting and are
combined. In Frequency domain filtering, the fingerprint image is subdivided into 32*32 small frames. Features
are extracted from these frames in frequency domain. Final enhanced fingerprint is obtained the results of Gabor
filter and frequency domain filtering. Binarization and Thinning follows next where the enhanced fingerprint is
converted to binary and the ridges are thinned to one pixel width. This helps in extracting the Minutiae parts
(ridge bifurcation and ridge endings). Euclidian distance method used for performance of recognition. The
overall recognition rate for the proposed method is 95% obtained as which is much better compared to histogram
method where the recognition rate is 64%. This project is implemented using MATLAB.
Keywords - Gabor filter, fingerprint, FFT, Minutia, AFIS.

I.

Introduction

The biometry or biometrics refers to the automatic identification of an individual by using certain
physiological or behavioural traits associated with the person. Traditionally, passwords and ID cards have been
used to moderate access to restricted systems. However, security can be easily breached in these systems when a
password is divulged to an unauthorized user or an impostor steals a card. Furthermore, simple passwords are easy
to guess and difficult passwords may be hard to recall. Fingerprints are fully formed at about seven months of
fetus development and finger ridge configurations do not change throughout the life of an individual except due to
accidents such as bruises and cuts on the fingertips. This property makes fingerprints a very attractive biometric
identifier. Fingerprint recognition represents the oldest method of biometric identification. Its history is going
back as far as at least 2200 BC. Since 1897, has been used for criminal identification. A fingerprint consists of
ridges and valleys. The pattern of the ridges and valleys is unique for each individual. There are two major
methods of fingerprint matching: Local Features and global pattern matching. The first approach analyses ridge
bifurcations and endings, the second method represents a more macroscopic approach. The last approach
considers the flow of ridges in terms of, for example, arches loops and whorls. As the equal-error-rate is low,
therefore fingerprint recognition is very accurate. Another important characteristic is to take into account the type
from used reader, one of capacitive surface or of optical surface; we used of optical surface for this work. In an
image between greater is the quality, lesser is the probability of finding a false minutia. A false minutia is created
by a bad quality of the image.

II.

Gabor Filter

A Gabor filter is a linear filter whose impulse response is defined by a harmonic function multiplied by a
Gaussian function. Because of the multiplication-convolution property (Convolution theorem), the Fourier
transform of a Gabor filter's impulse response is the convolution of the Fourier transform of the harmonic function
and the Fourier transform of the Gaussian function.
A Gabor filter is a linear filter used for edge detection in image processing which is named after Dennis
Gabor. Gabor filter frequency and orientation representations are similar to those of human visual system, for
texture representation and discrimination it has been found to be remarkably appropriate. A sinusoidal plane wave
has been modulating a 2D Gabor filter which is a Gaussian kernel function in the spatial domain. From one parent
wavelet all filters can be generated by dilation and rotation, thus the Gabor filters are self-similar. With eight
different orientations of Gabor filter, features of the fingerprint are extracting and are combined. where f
represents the ridge frequency and the choice of 2 and 2 determines the shape of the filter envelope and also
the trade of between enhancement and spurious artifacts. This is by far, the most popular approach for fingerprint
enhancement
www.iosrjournals.org

17 | Page

Fingerprint Recognition Using Gabor Filter and Frequency Domain Filtering

1

G(x,y) = exp{- 2 [

III.

+ 2 ]} cos(2)

(1)

Frequency Domain Filtering

The main aim of enhancement to improve the clarity of ridge structure in the recoverable region and
mark unrecoverable regions
We divide the image into small processing blocks (32 by 32 pixels) and perform the Fourier transform according
to:
ux
vy
N1
F U, V = M1
(2)
x=0
y=0 f(x, y) exp j2 M + N
For u = 0, 1, 2, ... 31 and v = 0, 1, 2, ... 31
In order to enhance a specific block by its dominant frequencies, we multiply the FFT of the block by its
magnitude a set of times. Where the magnitude of the original
FFT=abs(F(u,v))=|F(u,v)|.

IV.

Proposed Enhancement

From the previous subsections is clear that both approaches present desirable features that can be
combined to obtain better image enhancement results. Thus this project proposes to use a combination of Fourier
transform and Gabor filtering to carry out the image enhancement task. In the final image with minutiaes detected
in each stage is observed. Since we have the two enhanced images an algebraic sum is made and only the resulting
pixel will be white, if in the two images the pixel is white too.
Image
Enhanced
with
Gabor
filter

im

Im(i,
j)>2
55

ADD

Y
Im(i,j)=2
55

Final image

N
Image
Enhanced
with FFT

N
Im(i,j)=0

Fig.1. Block diagram of proposed fingerprint enhancement method

0

V.

Binarization

Image binarization is the process of turning a gray scale image to a black and white image. In a grayscale image, a pixel can take on 256 different intensity values while each pixel is assigned to be either black or
white in a black and white image. This conversion from gray-scale to black and white is performed by applying a
threshold value to the image. In MATLAB, a value of one means the pixel is white, whereas a value of zero
indicates the pixel is black. For a gray-scale image, the pixels are decimal values between zero and one. When a
threshold is applied to an image, all pixel values are compared to the input threshold. Any pixel values below the
threshold are set to zero, and any values greater than the threshold are set to one. By the end of this process, all
pixel values within the image are either zero or one, and the image has been converted to binary format.

VI.

Thinning

Ridge Thinning is to eliminate the redundant pixels of ridges till the ridges are just one pixel wide. In
each scan of the full fingerprint image, the algorithm marks down redundant pixels in each small image window
(3x3). And finally removes all those marked pixels after several scans. In my testing, such an iterative, parallel
thinning algorithm has bad efficiency although it can get an ideal thinned ridge map after enough scans. Their
method traces along the ridges having maximum gray intensity value. However, binarization is implicitly enforced
since only pixels with maximum gray intensity value are remained. Also in my testing, the advancement of each
trace step still has large computation complexity although it does not require the movement of pixel by pixel as in
other thinning algorithms. This is done by using the following MATLABs thinning function.

www.iosrjournals.org

18 | Page

Fingerprint Recognition Using Gabor Filter and Frequency Domain Filtering

VII.
Minutia Detection
After the fingerprint ridge thinning, marking minutia points is relatively easy. But it is still not a
trivial task as most literatures declared because at least one special case evokes my caution during the minutia
marking stage. In general, for each 3x3 window, if the central pixel is 1 and has exactly 3 one-value neighbours,
then the central pixel is a ridge branch. If the central pixel is 1 and has only 1 one-value neighbour, then the
central pixel is a ridge ending.

VIII.

Implementation

Template 1
False
recognition
Template .2

.
.
Image database

Comparison of ridge
endings, bifurcations,
Angles.

Template N

Decision based
on % matching

True
recognition
Test image

Test template

Fig.2. Implementation Block diagram

The block diagram which is used for fingerprint recognition is shown in the above figure. Image database
consists of number of fingerprint images. Each fingerprint creates their own templates. The templates consists of
ridge endings, ridge bifurcations and angle of orientations. Each fingerprint creates their own templates. Test
image is nothing but the fingerprint image that is used for recognition. The test image can also create their own
test template. Test template consists of ridge endings, ridge bifurcations and angle of orientations. Ridge endings
and ridge bifurcations consists of X and Y coordinates. Then we can compare the edges, bifurcations, angles of
image database templates and test templates.

IX.

Fig.3. input image

Results

Fig.4. Gabor filter enhancement

www.iosrjournals.org

19 | Page

Fingerprint Recognition Using Gabor Filter and Frequency Domain Filtering

Fig.5. FFT enhancement

Fig.6. Proposed enhancement

Fig.7. Binarized image

Fig.8. Thinned image

Fig.9. Minutia marking

Fig.10 False minutia removal

120
100

% matching

80

PROPOSED
METHOD

60

HISTOGRAM
METHOD

40
20
0
1

Fig.11. Percentage recognition comparison for proposed method and histogram method
www.iosrjournals.org

20 | Page

Fingerprint Recognition Using Gabor Filter and Frequency Domain Filtering

X.
Summary And Conclusion
This project Fingerprint recognition using Gabor filter and Frequency domain filtering involves the
following steps: collection of database, Gabor filter enhancement, FFT enhancement, Proposed enhancement,
Binarization, Thinning, Minutia detection.
The image database consists of eight fingers of nine different persons. So, the database consists of totally
seventy two fingers. It can also be downloaded from fingerprint verification combination 2002(FVC 2002). With
eight different orientations of Gabor filter, features of the fingerprint are extracting and are combined. In
Frequency domain filtering, the fingerprint image is subdivided into 32*32 small frames. Features are extracted
from these frames in frequency domain. In this project we used combination of Gabor filter and FFT
enhancements. For the resultant enhanced image binarization[21] technique was applied in which black pixels are
assigned to 0, white pixels are assigned to 1. Thinning was applied to the binarized image which produces the
Skeleton structure [2]. Then we mark the minutias, by using this minutia markings the recognition was done [3].
The overall recognition rate for the proposed method is 95% obtained as which is much better
compared to histogram method where the recognition rate is 64%. This project is implemented using MATLAB.

XI.

Conclusion

From the reported investigations, it can be concluded that the overall recognition rate is higher for the
proposed method compared to the histogram method. In this project we use Euclidian distance method for
fingerprint recognition. Usually the reason for fault recognition or low recognition rate occurs due to positioning
of fingerprint. For example fingerprints in the database or the ones used for testing might be oriented in a
particular angle or might be positioned for left or right, top or bottom of the image. Due to this, fingerprint
matching might get faulty. One way to reduce the fault recognition is to reorient or readjust the fingerprints such
that the core comes at the center of the image and then start the matching process[8]. This can be scope for future
improvement of the project.

REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]

Leung, M. Engeler, and Frank, Fingerprint image processing using neural network. Proceedings of the IEEE conf. on Computer and
CommunicationSystems, pp. 671-774, 1990.
Louisa Lam,Seong Whan Lee Thinning Methodologies A ComprehensiveSurvey, IEEE Transactions on Pattern Analysis and
Machine Intelligence, Vol.14, no. 9, pp. 869-885, 1992.
Dario Maio and Davide Maltoni Direct Gray-Scale Minutiae Detection in Fingerprints, IEEE Transactions on pattern Analysis and
Machine Intelligence, Vol. 19, no. 1, pp. 27-40, 1997.
Ton van der Putte and Jeroen Keuning Biometrical Fingerprint Recognition, Fourth Working Conference on Smart Card Research
and Advanced Applications, pp 289-303, 2000.
Zsolt Miklo, Kova CS Vajna A Fingerprint Verification System Based onTriangular Matching and Dynamic Time Warping,
IEEE Transactions on PatternAnalysis and machine Intelligence, Vol. 22, NO. 11, pp. 1266-1276, 2000.
A.K.Jain, R.Bolle, and S.Pankanti, Biometrics Personal Identification inNetworked Society, Kluwer Academic Publishers New
York, Boston, Dordrecht,London, Moscow, pp. 1-64, 2002.
Bolle, R.,Serior, A.Ratha, and Pankanti, Fingerprint minutiae a constructivedefinition. Proceedings of the workshop on Biometric
Authentication, PP, 5866(2002)
Milene Arantes, Alessandro Noriaki Ide A System ForFingerprint MinuteClassification and Recognition Proceedings of the IEEE
International Conference on Neural Information Processing, Vol. 5, pp. 2474-2478, 2002.

www.iosrjournals.org

21 | Page