International Conference on Global Trends in Engineering, Technology and Management (ICGTETM-2016)
The Fingerprint Enhancement Techniques – Study and
Review
Mayur S. Patil1, Sandip S. Patil2
1
PG Student, Department of Computer Engineering, North Maharashtra University, Jalgaon
SSBT’s COET, Bambhori, Jalgaon, Maharashtra, India
2
Associate Professor, Department of Computer Engineering, North Maharashtra University, Jalgaon
SSBT’s COET, Bambhori, Jalgaon, Maharashtra, India
Abstract: Now a days, the biometric identification is
most accepted. It is well known that human being has
a unique fingerprint pattern which does not change
over the whole life time. The fingerprints of even
identical twins are different. That’s why the
fingerprints are so much popular as the biometric
identifier. In this the main thing is to match the
original fingerprint with the existing fingerprint which
is present in the datasets. The image acquired for the
matching must be of fine quality and it must be
without of any kind of noise. The less the noise in the
fingerprint images the more accurate further
operations. However it is not always easy to obtain a
good quality of fingerprint. Because of this the
fingerprint image gives improper minutiae results.
Therefore the fingerprints must be enhanced to extract
the minutiae and get all the features of the
fingerprints. So that it can reduce the false recognition
or true rejection rates w.r.t. original fingerprints.
There are three main techniques of enhancement.
Pixel wise Enhancement Techniques, Contextual
Filter Enhancement Techniques and Multi Resolution
Enhancement Techniques. This paper focuses on these
various Fingerprint Enhancement Techniques.
shared or stolen. Also, they cannot differentiate
between unauthorized and authorized client.
Biometrics such as fingerprint, voice, face, etc.
offers way of reliable personal authentication that can
address these problems and is gaining citizen and
government acceptance. Reliable extraction of
features from poor quality fingerprint is the most
challenging problem faced in the area of fingerprint
recognition. Fingerprint image enhancement is the
first step in every Automatic Fingerprint Identification
System (AFIS), which improves the quality of the
fingerprint image by removing noise and blur, thereby
escalating the reliability of fingerprint recognition.
The following Figure 1 represents the various patterns
of Fingerprint.
Keywords: Biometric identification, minutiae,
fingerprint enhancement, Pixel wise Enhancement,
Contextual Filter, Multi Resolution Enhancement, etc.
I. INTRODUCTION
Figure 1: Pattern of Fingerprint
In this ever more digital globe, unfailing personal
authentication has become an essential human
computer interface activity. E-commerce, access to
computer networks and National Security are some
examples where establishing a person’s identity is
very important. Along with security variables [like
password, key, fingerprint, etc.], the fingerprint is the
most widely used entity since they are unique and do
not change over life time and even not at all need to be
remembered.[4]
Security measures apart from
biometric rely on knowledge-based approaches like
passwords or token based approaches such as swipe
cards to control access to physical and virtual spaces.
Though ubiquitous, such methods are not very secure.
Tokens such as badges and access cards may be
ISSN: 2231-5381
II. BACKGROUND
An impression left by friction ridges on a surface is
known as fingerprint. If it is taken by direct imaging
(e.g. by putting the finger directly on a scanner), it is
called an online fingerprint whereas if the impression
is taken on any other surface, then it is called offline
fingerprint. A fingerprint pattern consists of ridges and
valleys. A ridge is a narrow elevated skin which looks
like a line on a finger surface and the valleys are the
furrows which separate the ridges. The neighbouring
page Figure 2 shows the general fingerprint
identification system.
http://www.ijettjournal.org
Page 630
International Conference on Global Trends in Engineering, Technology and Management (ICGTETM-2016)
III. MOTIVATION
Figure 2: General Fingerprint Identification System
There are three structural levels in a fingerprint
which are discussed below:
The motivation behind this is rising necessity to
identify an individual for security; fingerprint
enhancement to get better superiority of fingerprint
along with extract minutiae points. And in the
extraction process one should not get the bogus
minutiae and conserve the accurate ridge bifurcations
and endings. Minutia extracted from the fingerprint
greatly depends on quality of input finger print. For
extracting the exact Minutiae from the fingerprint it is
desirable to remove noise from the input image and
for that there is need of enhancement algorithm for
accuracy.
IV. FINGERPRINT ENHANCEMENT
TECHNIQUES
A. Global Pattern Level
This represents the overall pattern of a fingerprint.
For example, in a fingerprint, ridges may enter from
bottom-right and then after looping around the center,
they may exit from the bottom right again. According
to Henry [1] classification of fingerprints, it is called
right loop. Similarly, there are other such global
patterns and Henry’s classification in this respect is
exclusive and almost exhaustive. This structure is
mainly used for fingerprint classification and/or
indexing.
B. Local Ridge Pattern Level
This is the pattern on which most of the
identification algorithms are designed. Ridges in a
fingerprint may not be continuous. Some ridges split
into two ridges known as bifurcations while some end
at some point termed as ridge endings. Such positions
are important in a fingerprint. All these are called
minutiae points. Minutiae are points which are
combination of ridge ending and bifurcation. In
addition to this, there are some special points in a
fingerprint (called singular points), core point and
delta point. A core point is a point where the innermost ridge of a finger print turns. A delta point is a
point where two ridges running side-by-side diverge.
Not every fingerprint necessarily has core or delta
point while some may have more than one such point.
There are generally around 80 minutiae points in a
complete fingerprint of a finger tip.
C. Low Level Features
It consists of the sweat pores which appear as
white holes in the ridges. To obtain these features,
very high resolution scanners are required and so these
features are used by very sophisticated fingerprint
matching systems.
ISSN: 2231-5381
Image enhancement is the process of digitally
manipulating a stored image using software. The tools
used for image enhancement include many different
kinds of software such as filters, image editors and
other tools for changing various properties of an entire
image or parts of an image [2].
The main three types of Fingerprint enhancement
techniques having two or more subtypes of each are
discussed as follows [1]:
A. Pixel-wise enhancement
i.
Histogram Equalization
ii.
Contrast Enhancement
iii.
Normalization
iv.
Wiener Filter, etc.
In this pixel based fingerprint image processing
function the fresh value of each pixel only depends on
earlier value and a few universal parameters (not
depend on the value of the adjacent pixels). Pixel
based techniques don’t create satisfying and ultimate
results for fingerprint enhancement. Still, contrast
stretching, histogram manipulation, normalization and
Wiener filtering had given away to be helpful as initial
processing steps in a added elegant fingerprint
enhancement algorithm.
B. Contextual filtering
i.
Fourier Transform
ii.
Gabor Filter
iii.
Bell Shaped Filter, etc.
The broadly used method for fingerprint
enhancement is based on contextual filter. In
traditional image filtering just single filter is used for
involvedness all the way through the fingerprint image.
Within contextual filtering technique, the filter
properties modify based on local context.
http://www.ijettjournal.org
Page 631
International Conference on Global Trends in Engineering, Technology and Management (ICGTETM-2016)
Frequently, a bunch of filters is pre-computed and
one of them is chosen for each fingerprint image area.
In fingerprint image enhancement, the context is often
defined by the local ridge orientation, local ridge
frequency. In reality the sinusoidal-shaped wave of
ridges and valleys is principally defined by a local
orientation, frequency that varies slowly across the
fingerprint area. A suitable filter that is tuned to the
home ridge frequency and direction can efficiently
removes the unwanted noise and preserve the accurate
ridge valley arrangement.
method to find out, ridge present on a fingerprint
image by directly scanned fingerprints or inked
impression.
Robert Hastings, in [8] developed a way for
enhancing the ridge pattern by using method of
oriented diffusion by adaptation of anisotropic
diffusion to smooth fingerprint in the direction parallel
to the ridge flow. Fingerprint image intensity differs
slickly as one traverse along the ridges or valleys by
removing most of little irregularities and breaks but
with the identity of individual ridges, valleys
conserved.
C. Multi Resolution Enhancement
i.
Dimensional Filter
ii.
Rotational Filter, etc.
R. Sonawane et al., in [9] given a technique by
introducing a special domain fingerprint image
enhancement way which decomposes the image into a
Multi-resolution enhancement has been planned [3] set of filtered fingerprint images after that orientation
to remove noise from fingerprints. Decomposing the field estimated. Superiority mask distinguishes the
image into various sub-images allows to pay damages recoverable as well the unrecoverable corrupted
for different noise components at various scales: regions from input fingerprint images are generated
particularly, at higher levels (low and middle using estimated orientation field, Input fingerprint
frequency bands) the rough ridge-valley flow is adaptively enhanced into recoverable regions.
cleaned and gaps are closed, whereas at the lower
Eric Kukula et al., in [10] proposed a way for
levels (upper frequencies) the improved details are
conserved. The enhanced image bands are then investigating the result of five various force levels on
fingerprint matching performance, fingerprint image
recombined to gain the ultimate fingerprint image.
superiority scores, minutiae count up between optical
as well as capacitance fingerprint sensors. Three
V. LITERATURE SURVEY
fingerprint images were taken together from right
index fingers of seventy five participants in favor of
Arun et al., in [4] given that, contrast
each sensing technology. Results disclose a important
enhancements
improve
the
perceptibility
differentiation in fingerprint image quality score based
(recognizable) of objects in the image by enhancing
on force level and every sensor technology, yet there
the brightness difference between objects and their
is no major differentiation in minutiae count up based
backgrounds. Contrast enhancements are normally
on force level of capacitance sensor. Fingerprint
performed since an contrast stretch followed with a
image quality score, shown affected by force and
tonal enhancement, even though these both can be
sensor kind is one of many factors which influence the
performed in single step. Contrast stretch improves the
system matching performance, yet removal of low
brightness differences uniformly across the dynamic
quality fingerprint images doesn’t improve the system
range of the image.
performance at every force level.
Prasanna et al., in [5] the challenges and the
opportunities regarding the wet fingerprint recognition
are given. Also authors have introduced a new
database called WWF viz. Wet and Wrinkled
Fingerprint.
Shlomo et al., in [6] author proposed a pixel-wise
adaptive Wiener method for noise reduction. Filter
was based on local data estimated from a local
neighborhood of size 3 X 3 of each pixel.
Er.Nishi et al., in [3] in this paper in equalization
method it improves the appearance and spread the
quality of gray levels in order that they are equally
distribute across their range.
G. Sambasiva et al., in [7] projected fingerprint
identification technique using a gray level watershed
ISSN: 2231-5381
L. Lam et al., in [11] presented a method, thinning
is the process of reducing thickness of each line of
patterns to just an only one pixel width. Requirements
of an excellent algorithm w.r.t a fingerprint are i)
Thinned fingerprint image obtained should be of only
one pixel width with no discontinuities ii) Every ridge
must be thinned to this central pixel iii) Singular pixel,
noise should be eliminated iv) No extra removal of
pixel should be possible after completion of thinning
process.
Mana Tarjoman et al., in [12] introduced structural
approach to fingerprint classifications by using the
directional image of fingerprint instead of singularities.
Directional finferprint image includes dominant
direction of ridge lines.
http://www.ijettjournal.org
Page 632
International Conference on Global Trends in Engineering, Technology and Management (ICGTETM-2016)
Bhupesh Gour et al., in [13] have developed a
method for extraction of minutiae from fingerprint
images using midpoint ridge contour representation.
First step is segmentation to split foreground from
background of fingerprint image. The grayscale
intensities in 64x64 regions are normalized to a
constant mean and variance to remove the effects of
sensor noise and grayscale variations due to finger
pressure differences. After the normalization the
contrast of the ridges are enhanced by filtering 64x64
normalized windows by appropriately tuned Gabor
filter. Processed image is after that scanned from top
to bottom and left to right and transition from white
(background) to black (foreground) are detected. The
proposed method takes less and do not detect any false
minutiae.
Hartwing Fronthaler et al., in [14] proposed
fingerprint enhancement to improve the matching
performance and computational efficiency by using an
image scale pyramid and directional filtering in the
spatial domain.
[9]
[10]
[11]
[12]
[13]
[14]
Digital Image Computing Techniques and Applications, pp.
245-252, 2007.
Raju Sonavane and B. S. Sawant “Noisy Fingerprint Image
Enhancement Technique for Image Analysis: A Structure
Similarity Measure Approach”, Journal of Computer Science
and Network Security, vol. 7 no. 9, pp. 225-230, 2007.
Eric P. Kukula, Christine R. Blomeke, Shimon K. Modi, and
Tephen J. Elliott “Effect of Human Interaction on Fingerprint
Matching Performance, Image Quality, and Minutiae Count”,
International Conference on Information Technology and
Applications, pp. 771-776, 2008.
L. Lam S. W. Lee, C. Y. Suen “Thinning Methodologies-A
Comprehensive Survey”, IEEE Transactions on Pattern
Analysis and Machine Intelligence, vol. 14, pp. 869-885,
1992.
Mana Tarjoman, and Shaghayegh Zarei “Automatic
Fingerprint
Classification
using
Graph
Theory”,
Proceedings of World Academy of Science, Engineering and
Technology, vol. 30, pp. 831-835, 2008.
Bhupesh Gour, T. K. Bandopadhyaya and Sudhir Sharma
“Fingerprint Feature Extraction using Midpoint Ridge
Contour Method and Neural Network”, International Journal
of Computer Science and Network Security, vol. 8, no, 7, pp.
99-109, 2008.
Hartwing Fronthaler, Klaus kollreider, and Josef Bigun
“Local Features for Enhancement and Minutiae Extraction
in Fingerprints”, IEEE Transactions on Image Processing,
vol. 17, no, 3, pp. 354-363, 2008.
VI. CONCLUSION
This study and review of Fingerprint Enhancement
can be conclude that biometric identification is having
very much importance in today’s digital world. The
Pixel wise enhancement techniques are having very
basic and older functionalities. While the Contextual
fingerprint enhancement techniques are quite superior
to the previous one. And they are widely used.
In future also the most advanced techniques of the
fingerprint enhancement that is Multi resolution
fingerprint enhancement will come into picture and it
will produce constructive results.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Davide Maltoni, Dario Maio, Anil K. Jain, Salil Prabhakar
“Handbook of Fingerprint Recognition”, Second Edition,
Springer, 2009.
https://www.techopedia.com/de_nition/26314/imageenhancement
Er.Nishi Madaan, Er.Arun Begill, “A Comprehensive Review
of Various Image Enhancement Techniques”, International
Journal of Advanced Research in Computer Engineering
Technology, pp.1181-1185, April 2014.
Arun Kavi Arasu, Mohamed Nizar, Prabhakaran D, “Review
of Image Contrast Enhancement Techniques”, International
Journal of Engineering Research Technology, pp.473-480,
November 2013.
Prasanna Krishnasamy, Serge Belongie, David Kriegman
“Wet
Fingerprint
Recogniotion:
Challenges
and
Opportunities”, IEEE Conference Publications, pp. 1-7 (9781-4577-1357-6), 2011.
Shlomo Greenberg, Mayer Aladjem and Daniel Kogan
“Fingerprint Image Enhancement using Filtering
Techniques”, Elsevier Science Ltd, pp.226-236, 2002.
G.Sambasiva Rao, C. NagaRaju, L. S. S. Reddy and E. V.
Prasad “A Novel Fingerprints Identification System Based on
the Edge Detection”, International Journal of Computer
Science and Network Security, vol. 8, pp. 394-397, 2008.
Robert Hastings “Ridge Enhancement in Fingerprint Images
Using Oriented Diffusion”, IEEE Computer Society on
ISSN: 2231-5381
http://www.ijettjournal.org
Page 633