American Journal of Science and Technology
2015; 2(6): 262-269
Published online September 30, 2015 (http://www.aascit.org/journal/ajst)
ISSN: 2375-3846
Keywords Fingerprint,
Minutiae Extraction,
Minutiae Matching,
Binarization,
Smoothing,
Image Segmentation,
Matrix Equalization and
Bifurcation
Received: May 5, 2015
Revised: June 22, 2015
Accepted: June 23, 2015
Fingerprint Matching Through Minutiae Based Feature Extraction Method
Md. Shahadat Hossain1, Md. Rafiqul Islam
2
1Applied mathematics, Mathematics Discipline, Khulna University, Khulna, Bangladesh 2Mathematics Discipline, Khulna University, Khulna, Bangladesh
Email address [email protected] (Md. S. Hossain), [email protected] (Md. R. Islam)
Citation Md. Shahadat Hossain, Md. Rafiqul Islam. Fingerprint Matching Through Minutiae Based Feature
Extraction Method. American Journal of Science and Technology.
Vol. 2, No. 6, 2015, pp. 262-269.
Abstract Here minutiae based feature extraction method has been discussed which is used for
fingerprint matching. This method is mainly depending on the characteristics of minutiae
of the individuals. The minutiae are ridge endings or bifurcations on the fingerprints. Their
coordinates and direction are most distinctive features to represent the fingerprint. Most
fingerprint matching systems store only the minutiae template in the database for further
usage. The conventional methods to utilize minutiae information are treating it as a point
set and finding the matched points from different minutiae sets. This kind of
minutiae-based fingerprint recognition/matching systems consists of two steps: minutiae
extraction and minutiae matching. Image enhancement, histogram equalization, thinning,
binarization, smoothing, block direction estimation, image segmentation, ROI extraction
etc. are discussed in the minutiae extraction step. After the extraction of minutiae the false
minutiae are removed from the extraction to get the accurate result. In the minutiae
matching process, the minutiae features of a given fingerprint are compared with the
minutiae template and the matched minutiae will be found out. The final template used for
fingerprint matching is further utilized in the matching stage to enhance the system’s
performance.
1. Introduction
Fingerprint matching is a widely used biometric authentication system that is done on
the basis that every person in the world has its own fingerprint. Every fingerprint has its
universal and uniqueness characteristics and widely acceptability. We have mentioned that
each fingerprint is formed in the womb during the age 17th
week of the fetus and remain
unchanged throughout the whole life [4]. In our study we have tried to find out the
difference between the input fingerprints by concept of the uniqueness of the fingerprints.
For experiment we have taken the concept of matching any two fingerprints from the
online journal [3]. We have seen that their input fingerprint, histogram equalization for
enhancing the fingerprints, binarization, image segmentation (Block direction estimation
and ROI extraction), and false minutiae removal, minutia matcher (Alignment stage and
matching stage) are discussed. After that we have mentioned an exceptional concept that
every image give its own matrix which help to find out the difference between the input
fingerprints. In the next stage we have seen although each fingerprints have a bifurcation,
in the first fingerprint image there are two minutiae in the same direction where in the
second fingerprint image the rest two minutiae are in the different direction. In our study
we have taken the help of MATLAB programming and at one stage the AFIS (Automated
263 Md. Shahadat Hossain and Md. Rafiqul Islam: Fingerprint Matching Through Minutiae Based Feature Extraction Method
Fingerprint Identification System) software for the
implementation of the matching of two different fingerprint
images.
1.1. Minutiae Features
The major minutiae 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. Minutiae and patterns
are very important in the analysis of fingerprints since no two
fingers have been shown to be identical.[6]
Figure 1. Bifurcation and ridge ending.
Figure 2. Short ridge (dot).
1.2. Minutiae Extraction
Typically each detected minutiae is described by four
parameters:
where:
– are coordinates of the minutiae point,
– is minutiae direction typically obtained from local
ridge orientation,
– is type of the minutiae point (ridge ending or ridge
bifurcation),
The position of the minutiae point is at the tip of the ridge or
the valley and the direction is computed to the X axis (Fig 1).
2. Methodology
2.1. Fingerprint Matching Algorithm
The algorithm for matching two fingerprint images are
mentioned below
Figure 3. The fingerprint matching procedure.
2.2. Input Images & Histogram Equalization
Histogram equalization is widely used for contrast
enhancement in a variety of applications due to its simple
function [1]. Histeq performs histogram equalization in
MATLAB.
In histogram equalization the input pixel intensity x is
transformed to new intensity value . The transformed
function T is the product of a cumulative histogram and a scale
factor [11]. The scale factor is needed to fit the new intensity
value within the range of the intensity values, for example
Figure 4. Image-1 of a fingerprint and its histogram and equalization.
im
( ), , , ..................(i)i i i i im x y tθ=
,i ix y
iθ
it
byx T
0 255∼
American Journal of Science and Technology 2015; 2(6): 262-269 264
Figure 5. Image 2 of another fingerprint and its histogram and equalization.
2.3. Image Enhancement
Sometimes, the images have obtained does not have good
quality and so the quality of image can be upgraded by
enhancing the image, and thus the contrast between ridges and
valleys can be increased.
Figure 6. The enhanced image-1 after histogram equalization.
Figure 7. The enhanced image-2 after histogram equalization.
2.4. Edge Detection
Figure 8. Edge detected of image-1.
Figure 9. Edge detected of image-2.
The purpose of edge detection in AFIS is to significantly
reduce the amount of data found in a fingerprint image and
leave only the most important information. Edge detection
works by finding points on an image where the gray scale
value changes greatly between pixels [2].
2.5. Binarization of the Input Fingerprints
Some features of binarization is mentioned below
� Image binarization converts an image of up to 256 gray
levels to a black and white image. Frequently,
binarization is used as a pre-processor before OCR. In
fact, most OCR packages on the market work only on
bi-level (black & white) images.
� The simplest way to use image binarization is to choose
a threshold value, and classify all pixels with values
above this threshold as white, and all other pixels as
black. The problem then is how to select the correct
threshold. In many cases, finding one threshold
compatible to the entire image is very difficult, and in
many cases even impossible. Therefore, adaptive image
binarization is needed where an optimal threshold is
chosen for each image area.
265 Md. Shahadat Hossain and Md. Rafiqul Islam: Fingerprint Matching Through Minutiae Based Feature Extraction Method
Figure 10. The Binarized images of the fingerprint image-2.
Figure 11 .The Binarized images of the fingerprint image-1.
2.6. Image Segmentation
In general, only a Region of Interest (ROI) is useful to be
recognized for each fingerprint image. The image area without
effective ridges and furrows is first discarded since it only
holds background information. Then the bound of the
remaining effective area is sketched out since the minutiae in
the bound region are confusing with that spurious minutia that
is generated when the ridges are out of the sensor.
Image segmentation is classified into two major part
� Block direction estimation and
� ROI extraction
2.6.1. Block Direction Estimation
To estimate the block direction for each block of the
fingerprint image an algorithm is needed which is mentioned
below
1. At first it needs to calculate the gradient values along
both x-direction and y-direction for each
pixel of the block.
2. For each block, there needs to use following formula to
get the Least Square approximation of the block
direction.
for all the pixels in each block.
The formula mentioned above is easy to understand by
regarding gradient values along x-direction and y-direction as
cosine value and sine value. Therefore the tangent value of the
block direction can be estimated nearly the same as the way
illustrated by the following formula.
After completing the estimation of each block direction,
those blocks without having significant information on ridges
and furrows are discarded based on the following formulas:
For each block, if its certainty level E is below a threshold,
then the block is regarded as a background block.
Figure 12. Block direction of the fingerprint-1.
Figure 13. Block direction of the fingerprint-2.
2.6.2. ROI Extraction (Morphological Method)
Close (shrink images and eliminate small cavities)
Open (expands images and remove peaks introduced by
background noise)
( )xg ( )yg
( )( )2 2
2tan 2
x y
x y
g g
g gβ =
−∑∑
∑∑
2 2
2sin costan 2
cos sin
θ θθθ θ
=−
( ) ( )( )
2 2
2 2
2 x y x y
x y
g g g gE
WW g g
+ −=
+∑∑ ∑∑
∑∑
American Journal of Science and Technology 2015; 2(6): 262-269 266
Figure 14. Image of the fingerprint 1.
Figure 15. Image of the fingerprint 2.
2.7. Thinning of the Input Fingerprint
Thinning is defined as a procedure to transform a digital
binary pattern to a connected skeleton of unit width. After
thinning the minutiae of the original fingerprint become more
visible in the thinned fingerprint image. And this thinned
fingerprint image is used to match with the other fingerprint
images so that the variation of features among the
fingerprints can be detected easily.
Figure 16. The thinned image-1.
Figure 17. The thinned image- 2.
2.8. Termination and Bifurcation
Since various data acquisition conditions such as
impression pressure can easily change one type of minutia into
the other, most researchers adopt the unification
representation for both termination and bifurcation. So each
minutia is completely characterized by the following
parameters at last: 1) x-coordinate, 2) y-coordinate, and 3)
orientation. The orientation calculation for a bifurcation needs
to be specially considered. All three ridges deriving from the
bifurcation point have their own direction represents the
bifurcation orientation using a technique proposed in [7]
simply chooses the minimum angle among the three
anticlockwise orientations starting from the x-axis. Both
methods cast the other two directions away, so some
information loses. The termination and bifurcation of both
fingerprints are mentioned below.
Figure 18. Termination of 1.
267 Md. Shahadat Hossain and Md. Rafiqul Islam: Fingerprint Matching Through Minutiae Based Feature Extraction Method
Figure 19. Termination of 2.
Figure 20. Bifurcation of image-1.
Figure 21. Bifurcation of image -2.
Since various data acquisition conditions such as
impression pressure can easily change one type of minutia into
the other, most researchers adopt the unification
representation for both termination and bifurcation.
2.9. Removal of the False Minutiae
The false minutiae can affect the result of fingerprint
matching. These types of minutia are to be removed.
Figure 22. Image-1 after removal of false minutiae.
Figure 23. Image-2 after removal of false minutiae.
2.10. Minutiae in the ROI of the Fingerprints
In this stage marked minutiae are highlighted with different
colors so that it can be found out the difference between any
two fingerprints.
Figure 24. Image of fingerprint 1.
American Journal of Science and Technology 2015; 2(6): 262-269 268
Figure 25. Image of fingerprint 2.
2.11. Unique Minutiae Sorter
The matching of the fingerprint includes some procedures
which are mentioned through the following figures
Figure 26. Image of the fingerprint 1.
Figure 27. Image of the fingerprint 2.
2.12. Minutiae Matching
Feature-based(Minutiae-based) Matching: Typical
fingerprint recognition methods employ feature-based
matching, where minutiae (i.e., ridge ending and ridge
bifurcation) are extracted from the registered fingerprint
image and the input fingerprint image, and the number of
corresponding minutiae pairings between the two images is
used to recognize a valid fingerprint image. Alternatively, Jain
et al. [8] used a string matching technique while Isenor and
Zaky [9] propose a graph-based fingerprint matching
algorithm. In are [10] describes a fingerprint verification
algorithm based on a bipartite graph construction between
model and query fingerprint feature clusters. The minutiae
matching problem has been generally addressed as a point
pattern matching problem which has been extensively studied
yielding families of approaches known as relaxation methods,
algebraic and operational research solutions, tree-pruning
approaches, energy- minimization methods, Hough transform,
etc.
3. Experimental Results and Discussions
From the tree diagram mentioned above it is notable that the
tree diagram of the two given fingerprints are totally different
from each other. So the given two fingerprints are also
different from each other.
Figure 28. Direction of minutia of image 1.
Figure 29. Direction of minutia of image 2.
From the above figure-28 and figure-29 it is found that in
269 Md. Shahadat Hossain and Md. Rafiqul Islam: Fingerprint Matching Through Minutiae Based Feature Extraction Method
each fingerprint there is one bifurcation. But in the first image
the other two minutiae are in the same direction while in the
second one image the other two minutiae are in the different
direction. So it can be reached to the decision that the two
fingerprints are totally different.
4. Conclusions
Our vision has combined a method to build a minutia
extractor and a minutia matcher. The combination of multiple
methods comes from a wide investigation into different
research papers. Also some significant changes like
segmentation using Morphological operations, minutia
marking with special considering the triple branch counting,
minutia unification by decomposing a branch into three
terminations and matching in the unified x-y coordinate
system after a two-step transformation are used in our project,
which are not reported in other literatures we referred to. The
processes named thinning, binarization, smoothing , block
direction estimation, absolute contrast, image segmentation,
ROI extraction etc. are done by using the update software
“SourceAFIS-1.7.0” where AFIS means Automated
Fingerprint Identification System .Also a program coding
with MATLAB going through all the stages of the fingerprint
matching is built. It is helpful to understand the procedures of
fingerprint matching. And demonstrate the key issues of
fingerprint matching. At last by the Matlab code for
equalizing two matrices of the two input fingerprints it is
found that the result is zero which means that the two
fingerprints taken as input are absolutely different from each
other.
Acknowledgement
I would like to give thank Professor Dr. Md. Rafiqul Islam
Sir for his time to time, very much needed, valuable guidance.
I am also grateful to him who encouraged me to make this
effort a success.
References
[1] Bassiou, N. and Kotropoulos, C., "Color image histogram equalization by absolute discounting back-off," Computer Vision and Image Understanding, 107(1-2):108-122,
[2] Boldischar, M. and Moua, C. P., “Edge Detection and Feature Extraction in Automated Fingerprint Identification Systems”
[3] Applications of fingerprint matching are available at http://www.answers.com/Q/What_are_the_practical_applications_of_fingerprinting
[4] Lasting impression of fingerprint is available at http://www.livescience.com/ 30- lasting-impression-fingerprints-created.html
[5] Image binarization is available at https://www.research.ibm.com/haifa/projects/image/glt/binar.html
[6] Thornton, John (May 9, 2000). Latent Fingerprints, Setting Standards In The Comparison and Identification. 84th Annual Training Conference of the California State Division of IAI. Retrieved 30 August 2010.
[7] Applications of fingerprint matching are available at http://www.answers.com/Q/What_are_the_practical_applications_of_fingerprinting
[8] A. K. Jain, L. Hong, R. M. Bolle, "On-line fingerprint verification", IEEE Trans, on Pattern Analysis and Machine Intelligence, Vol. 19, No 4, pp.302-313, 1997.
[9] D. K. Isenor, S. G. Zaky, "Fingerprint identification using graph matching", Pattern Recognition, Vol. 19, No 2, pp. 113-122, 1986.
[10] K.-C. Fan, C.-W. Liu, Y.-K. Wang, "A fuzzy bipartite weighted graph matching approach to fingerprint verification", In Proc. of the IEEE International Conf. on Systems, Man and Cybernetics, pp. 729-733, Oct 1998.
[11] Formula of histogram equalization is available at http://www.songho.ca/dsp/histogram/histogram.html
Biography
Md. Shahadat Hossain: I am Md. Shahadat Hossain. I was born in 1st January, 1992 in the village named Khuriakhali of Sharonkhola Upazilla under Bagerhat district. I have completed my bachelor degree in mathematics from Khulna university, Bangladesh in the year 2014. Now I am a student of M.Sc. of the same institution in applied mathematics. In my free time I am engaged in research related to mathematics.
Md. Rafiqul Islam: I am Dr. Md. Rafiqul Islam. I was born in 1966 in the district of Satkhira, Bangladesh. I have taken my bachelor degree in mathematics from University of Rajshahi. I have taken my M.Sc. degree in applied mathematics from University of Saudi Arabia. I have completed my Ph.D. degree in applied mathematics from University of Rajshahi. At present I am a professor of mathematics discipline, Khulna University, Bangladesh.