(IJCSIS) International Journal of Computer Science and Information Security,

Vol. 6, No. 2, 2009

Fingerprint Verification based on Gabor Filter

Enhancement
Lavanya B N*, K B Raja* and Venugopal K R*
*Department of Computer Science and Engineering
University Visvesvariah College of Engineering, Bangalore University, Bangalore 560 001
[email protected]

Abstract—Human fingerprints are reliable to investigate the images of fingerprint by using

characteristics for personnel identification as it is suitable filters. Quality of fingerprint is assessed by
unique and persistence. A fingerprint pattern Character, Fidelity and Utility.
consists of ridges, valleys and minutiae. In this paper
we propose Fingerprint Verification based on Gabor A smoothly flowing pattern formed by
Filter Enhancement (FVGFE) algorithm for alternating crests (ridges) and troughs (valleys) on
minutiae feature extraction and post processing the palmer aspect of hand is palm print. Formation
based on 9-pixel neighborhood. A global feature of a palm print depends on the initial conditions of
extraction and fingerprints enhancement are based the embryonic mesoderm from which they develop.
on Hong enhancement method which is The pattern on pulp of each terminal phalanx is
simultaneously able to extract local ridge orientation considered as an individual pattern and is
and ridge frequency. It is observed that the commonly referred to as a Fingerprint. A
Sensitivity and Specificity values are better fingerprint is believed to be unique to each person,
compared to the existing algorithms. even identical twins have different. It is one of the
most mature biometric technologies and is
Keywords-fingerprints; biometrics; ridge; considered legitimate proofs of evidence all over
orientation image; minutiae extraction. the world, hence used in forensic divisions
worldwide for criminal investigations. Fingerprint-
I. INTRODUCTION based identification has better matching
performance than any other existing biometric
The term Biometrics relates to the measurement technology. The watermark is embedded into the
(metric) of characteristics of a living (Bio) thing in region of interest in the Fingerprint image to
order to identify a person. Biometric recognition is enhance the security of Biometric System.
used as an automatic recognition of individuals
based on the physiological or behavioral An automatic fingerprint matching is popular in
characteristics. A physiological characteristic such systems which control access to physical locations,
as Fingerprint, Face, Iris, Hand geometry and computer/network resources, bank accounts or
Retina remains same throughout the lifetime of a register employee attendance time in company. The
person. Behavioral characteristics such as matching between the fingerprint pattern are highly
signature, gait, voice and keystroke changes with sensitive to errors such as various noises, damaged
age and mentality of a person. fingerprint areas, the finger being placed in
different areas of fingerprint scanner window with
Fingerprint is one of the most widely used different orientation angles and finger deformation
biometric because individuals have tried a number during the scanning procedure etc.
of methods to obliterate or remove their
fingerprints by abrading or trying to rub off their Minutiae in the fingerprint images are the
fingerprints, burning the fingers, trying to dissolve terminations and bifurcations of ridge lines. To
their fingerprints with strong acids, covering their achieve high accuracy in minutiae of fingerprint
fingertips with superglue, and even having doctors images various algorithm is used and they are (i)
transplant skin from one finger to another segmentation algorithm which separate fingerprint
unsuccessfully. The fingerprint is formed in third image from any noisy background without
and fourth month of foetal development and unique excluding any ridge valley regions, (ii) image
with epidermal ridges, furrows and patterns. enhancement algorithm preserves the original ridge
Distinctiveness and persistence are the highly flow pattern from altering, (iii) minutiae detection
desirable qualities of fingerprint. Factors such as algorithm which locates the minutiae points.
Skin condition, finger pressure and noise detroit the Challenge is made to show that the three levels of
quality of fingerprint images. The presence of information about the parent fingerprint can be
noise gives an image a mottled, grainy, textured or elicited from the minutiae template alone, i.e. the
snowy appearance. The images of fingerprint are orientation field information, the class or type
mistaken during termination; hence it is necessary information and the friction ridge structure.

138 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

Experiments using commercial fingerprint matcher five major classes of fingerprints are whorl, right
suggest that the reconstructed ridge structure bears loop, left loop, arch, and tented arch as shown in
close resemblance to the parent fingerprint. the Figure 1.
Fingerprint is represented as graph in which
minutiae acts as vertex set and binary minutiae Contribution: In this paper, the fingerprint
relations provide edge set. Transformation- image is enhancement by ridge orientation, ridge
invariant and transformation-variant features are frequency and Gabor Filter. Minutiae feature
extracted from binary relation. Local matching extraction is based on 9-pixel neighborhood
probability is estimated from transformation- method.
invariant features and fingerprint rotation Organization: The paper is organized into the
transformation with the adaptive Parson window is following sections. Section 2 is an overview of
modeled by transformation-variant features. related work. The FVGFE model is described in
Digital camera is used to present a touch-less Section 3. Section 4 discusses the algorithm for
fingerprint recognition system. Fingerprint images FVGFE system. Performance analysis of the
that are acquired using digital camera consist of system is presented in Section 5 and Conclusions
are contained in Section 6.
certain constraints such as low contrast between the
ridges and the valleys, defocus and motion
blurriness. Touch-less fingerprint recognition II. RELATED WORK
system comprises of preprocessing, feature Michael Kucken and A. C. Newell [1],
extraction and matching stages. Preprocessing discussed the hypothesis on the development of
comprises of promising results in terms of epidermal ridges viz., (i) the epidermal ridge
segmentation, enhancement and core point pattern is established as the result of buckling
detection. Feature extraction is done by Gabor filter instability acting on the basal layer of the epidermis
and matching by Support Vector Machine (SVM). and resulting in the primary ridges. (ii) The
buckling process underlying fingerprint
development is controlled by the stresses formed in
the basal layer and not by the curvatures of the skin
surface. (iii) The stresses that determine ridge
direction are themselves determined by boundary
forces acting at creases and nail furrow, normal
displacements which are most pronounced close to
the ridge.
(a) (b)
Daesung Moon et. at., [2] implemented the
fuzzy fingerprint vault system to secure critical
data with fingerprint data. Geometric Hashing and
RS decoding algorithms are implemented. Anil Jain
et. al., [3] have designed and implemented an on-
line fingerprint verification system which operates
in two stages: minutiae extraction and minutiae
matching. An alignment-based elastic matching
(c) (d) algorithm is proposed for minutiae matching. The
matching algorithm has the ability to adaptively
compensate for the nonlinear deformations and
inexact pose transformations between different
fingerprints.
Shlomo Greenberg et. al., [4] proposed two
methods for fingerprint image enhancement. The
first one is carried out using local histogram
equalization, Wiener filtering, and image
binarization. The second method use a unique
(e)
anisotropic filter for direct grayscale
Figure 1. Five classes of Fingerprints (a) Arch (b) Tented enhancement. Seifedine Kadry and Aziz Barbar [5]
Arch (c) Whorl (d) Left Loop (e) Right Loop. designed Mobile handheld devices which consist of
Fingerprint classification is a technique to a technique to solve the authenticity problem in
assign a fingerprint into one of the several pre- mobile communication. It resists any forge
specified types already established in the literature imposed by another party.
which can provide an indexing mechanism. The

139 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

Tabassam Nawaz et. at., [6] proposed a system G. Jagadeeswar Reddy et. at., [14] presented
that maintains the whole process of taking fingerprint denoising using both wavelet and
attendance and maintaining records. The system curvelet transforms. The search-rearrangement
obtains attendance electronically using sensor and method performs better than minutiae based
all records are saved on a computer server. Michael matching for fingerprint binary constraint graph
Kaas and A. Witkin [7] discussed fingerprint matching since it does not require implicit
oriented patterns to decompose into two parts: a alignment of two fingerprint images. K. Zebbiche
flow field, describing the direction of anisotropy, and F. Khelifi [15] presented biometric images as
and the residual pattern obtained by describing the one Region of interest that has the data processed
image in a coordinate system built, from the flow by most biometric based system. The schemes
field. proposed here consist of embedding the watermark
into region of interest in fingerprint images.
Asker M. Bazen and Gerez [8] presented Discrete Wavelet Transform and Discrete Fourier
methods for the estimation of a high resolution Transform is used in this algorithm.
directional field from fingerprints. The directional
field detects the singular points and the orientations Bhupesh Gour et. at., [16] introduced midpoint
of the points are obtained. ridge contour representation inorder to extract the
minutiae from fingerprint images. Colour coding
Lin Hong et. al., [9] introduced a fingerprint scheme is used to scan each ridge only once. K.
enhancement algorithm which decomposes the Karu et. al., [17] proposed fingerprint classification
input fingerprint image into a set of filtered images. algorithm. Here classification is based on the
The orientation field is estimated and a quality number and locations of the detected singular
mask distinguishes the recoverable and points. The algorithm consists of three major steps:
unrecoverable corrupted regions in the input image (i) computation of the ridge directions, (ii) finding
are generated. Using the estimated orientation field, the singularities in the directional image, and (iii)
the input fingerprint image is adaptively enhanced classification of the fingerprint based on the
in the recoverable regions. detected singular points.
B. G. Sherlock et. al., [10] described a method L.Hong et. at., [19] incorporated a fingerprint
of enhancing fingerprint images, based upon non enhancement algorithm in the minutiae extraction
stationary directional Fourier domain filtering. module. Fast fingerprint enhancement algorithm
Fingerprints are first smoothed using a directional can improve the clarity of ridge and furrow
filter whose orientation is everywhere matched to structures based on estimated local ridge
the local ridge orientation by threshold. orientation and frequency.
Yun and Cho [11] proposed an adaptive
preprocessing method, which extracts five features III. MODEL
from the fingerprint images, analyzes image quality
with clustering method, and enhances the images A. Definitions
according to their characteristics. The
preprocessing is performed after distinguishing the 1) Fingerprint: Impression of a finger acquired
fingerprint image quality according to its from digital scanners.
characteristics.
2) Minutiae: Ridge bifurcations and ridge
Brankica M. Popovi´c and L. Ma’skovi c [12] endings in fingerprint image.
used multiscale directional information obtained
3) Core: Centre of the loop or pattern in
from orientation field image to filter the spurious
minutiae. The goal of feature extraction in pattern fingerprint image. It is located where the
recognition system is to extract information from innermost recurve begins and curve to exit the
the input data and depends greatly on the quality of same way they came in.
the images. Multiscale directional information 4) Delta: It is the area of pattern where there is
estimation is based on orientation field estimation. a triangulation or a dividing of ridges.
5) Loop: The fingerprint image contains only
F. A. Afsar et. al., [13] presented the minutiae
based Automatic Fingerprint Identification one delta.
Systems. The technique is based on the extraction 6) Whorl: The fingerprint image contains 2 or
of minutiae from the thinned, binarized and more deltas.
segmented version of a fingerprint image. The 7) Sensitivity: The ability of the algorithm to
system uses fingerprint classification for indexing detect the true minutiae and is represented as, 1 –
during fingerprint matching which greatly enhances Missed minutiae / ground truth number minutiae.
the performance of the matching algorithm.

140 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

8) Specificity: The ability of the algorithm to 8 bit gray scale image can represent 2 8- 1 = 255
reject false minutiae and is represented as, 1 – intensity or gray levels. Most of the fingerprint
false minutiae / ground truth number minutiae. images are stored as 8 bit gray scale images
9) Spurious Minutiae: It is the type of error usually in a bitmap image or as a TIFF image. Due
to the large number of gray levels, processing
that falsely identifies a noisy ridge structure as
complexity increases. To overcome this, the image
minutiae. is converted into a Binary image.
10) Missed Minutiae: It is the type of error that
occurs in failing to detect the existing minutiae A binary image is stored as 1-bit image.
when it is obscured by surrounding noise, scars or Usually the grayscale image is converted into a
poor ridge structures. binary image using a threshold value. Let I(x, y)
represent the intensity value of the enhanced
11) False Minutiae: Points which have been
grayscale image at the pixel position (x, y).Let TP
incorrectly identified as minutiae. be the threshold value. In case of fingerprint
12) Orientation Image: The orientation Image images, TP represents the differentiating intensity
represents an intrinsic property of the fingerprint between the background pixels and the ridge pixels.
images and defines invariant coordinates for ridges BW(x, y) represent the binary image obtained by
and furrows in a local neighborhood by Least the Equation (1).
Mean Square orientation estimation algorithm
B. Block Diagram of FVGFE Fingerprint Image
Figure 2 gives the block diagram of FVGFE
to extract the Minutiae points for fingerprint
verification to disclose personnel identity. Enhancement
1) Fingerprint Image: A gray-level fingerprint
image (I) of size N*N is considered to test the
FVGFE algorithm. The fingerprint images are Binarization
scanned at a resolution of 500 dots per inch (dpi).
2) Fingerprint Image Enhancement:
Fingerprint Enhancement is used to improve the Thinning
image quality by removing noise by low-pass
filter. The configuration of parallel ridges and
furrows with well-defined frequency and Minutiae Extraction
orientation in a fingerprint image provide useful
information which helps in removing undesired
noise. The sinusoidal-shaped waves of ridges and
furrows vary slowly in a local constant orientation. Spurious Minutiae Removal
The band pass filter i.e. tuned to the corresponding
frequency and orientation can effectively remove
the undesired noise and preserve the true ridge and Figure 2. Block Diagram of FVGFE
furrow structures. The region mask is obtained by
classifying each block in the fingerprint image into
a recoverable or unrecoverable block and If the
percentage of recoverable regions is smaller than a
threshold, the fingerprint image is rejected. An
accepted image is then passed through the filtering
stage. Gabor filters have both frequency-selective (1)
and orientation-selective properties and have 4) Thinning: Thinning is achieved on the
optimal joint resolution in both spatial and binarized image by morphological operations. A
frequency domains. Therefore, it is appropriate to particular morphological operation called dilation
use Gabor filters as band pass filters to remove the is applied to the image until no further changes
noise and preserve true ridge/valley structures. occur. At this point, the fingerprint image is
3) Binarization: Binarization is the process thinned to a ridge thickness of 1- pixel width. The
where the enhanced grayscale image is converted advantages of thinning: Processing is faster,
into a binary image. A gray scale image is one Complexity is reduced, Memory requirements are
which has a specific number of gray levels. For an reduced and Execution time is less.

141 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

Using this transformation, the ridge pixels are

converted to white pixel and background is
converted to black pixels. The original fingerprint 1 1
image, Enhanced Image, Binarized Image and
Thinned Image are shown in the Figures 3 (a), X 1
3(b),3(c),3(d).
1

(a) (b)

Figure 5. Ridge End.

Figure 5 (a) shows a ridge ending with 2 pixels

in the 9 pixel neighborhood. Figure 5 (b) shows a
normal ridge pixel. All the ridge pixels are
(a) (b) characterized by 3 ones in the 9 – pixel
neighborhood to differentiate between ridge pixels
and minutiae points.

TABLE I. NUMBER OF PIXELS IN THE 9- PIXEL

NEIGHBORHOOD
Parameters
Ridges Ridge
Ridge end Ridge Pixel
bifurcation
Numbers 2 ≥4 3
(c) (d)
The number of pixels in the 9- pixel
Figure 3. (a) Input Image (b) Enhanced Image (c) neighborhood for Ridge End, Ridge Bifurcation
Binarized Image (d) Thinned Image. and Ridge Pixel are 2, 4 and 3 respectively shown
5) Minutiae extraction: Two types of minutiae in the Table I.
viz., Ridge bifurcation and Ridge ending are The redundant minutiae are reduced by (i) All
considered. The minutiae are extracted based on the minutiae points adjacent to each other and
the number of pixels in the 9 – pixel within a pre – specified window is removed. (ii) All
neighborhood. the minutiae points near the border and within a
certain fixed distance from it are removed. (iii) If
a) Ridge Bifurcation: In ridge bifurcation,
two ridge endings are encountered close to each
the number of pixels in the 9 pixel neighborhood is
other and if no ridges pass through them, and then
4 as shown in Figure 4.The X represents a valid they are reconnected. Spurious minutiae results
minutiae point and 1s in the 9 – pixel window from spikes and glitches in the thinned ridge are
represent the neighboring ridge pixels. Figure 4 (a) removed by (i) the distance from the ridge ending
and 4 (b) shows an example of vertical ridge to the ridge bifurcation is found. (ii) If the distance
bifurcation and horizontal ridge bifurcation is less than or equal to 6, then the minutiae is
respectively. removed.

1 1 1 IV. ALGORITHMS
1 X X A. Problem definition: Given a Fingerprint
1 Image, using FVGFE algorithm, minutiae
1
extraction is carried out for Fingerprint
verification.
(a) (b) B. The Objectives are:
Figure 4. Ridge bifurcation a) The filters are used to improve the quality
of fingerprint.
b) Ridge Ending: In ridge end, the number b) Binarization and Thinning are performed
of pixels in the 9 pixel neighborhood is 2. using threshold and dilation.

142 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

c) Minutiae are extracted. Table II gives the Mean and Standard Deviation
(SD) of sensitivity and specificity for existing
C. Assumptions: Fingerprint Extraction using Fourier Domain
a) Fingerprint database of fvc 2004 is Analysis (FEFD) and proposed FVGFE
considered algorithms. It is observed that the values of
Sensitivity and Specificity are more in the case of
b) Scanned fingerprints images are used.
proposed algorithm compared to the existing
Figure 6 shows the Algorithm of FVGFE for algorithm
Fingerprint verification.
VI. CONCLUSION
In this paper, we proposed FVGFE algorithm
• Input: Fingerprint image for extraction of minutiae. Fingerprint pattern is
• Output: Extracted Minutiae points enhancement to improve the clarity of the ridge
imposed in input image. and furrow structures by local ridge orientation
and ridge frequency. The minutiae points are
1) Fingerprint Enhancement is extracted from the fingerprint image using 9-pixel
performed to improve the image neighborhood method. The performance of
quality. proposed method is better than the existing method
in terms of sensitivity and specificity. In future the
2) Gray Scale image is converted to minutiae extraction algorithm is combined with
binary using Binarization. ridge extraction for better performance.

3) Ridge thickness of binarized Image REFERENCES

is reduced to 1-pixel width by [1] M. Kucken and A. C. Newell, “Fingerprint Formation,”
Thinning. Proceedings of Journal of Theoretical Biology, pp. 71–
83, 2005.
[2] Daesung Moon, Sungju Lee, Yongwha Chung,
4) Minutiae Extraction using 9-pixel “Implementation of Automatic Fuzzy Fingerprint Vault,”
neighborhood method. Proceedings of International conference on Machine
Learning and Cybernetics, pp. 3781-3786, July 2008.
5) Spurious Minutiae are removed [3] A. Jain, L. Hong and R. Bolle, “On-line Fingerprint
and true minutiae points are Verification,” IEEE-Pattern Analysis and Machine
Intelligence, vol.19, pp. 302-314, April. 1997.
imposed on image.
[4] S. Greenberg, M. Aladjem and D Kogan, “Fingerprint
Image Enhancement using Filtering Techniques,”
National Conference on Real-Time Imaging , pp. 227-
Figure 6: Algorithm of FVGFE 236, 2002.
[5] Seifedine Kadry, Aziz Barbar, “Design of Secure Mobile
V. PERFORMANCE ANALYSIS Communication using Fingerprint,” European Journal of
Scientific Research, vol. 30, pp.138-145, 2009.
For performance analysis, the Fingerprint
[6] Tabassam Nawaz, Saim Pervaiz, Arash Korrani,
database of FVC2004 (DB1) [18] is considered to “Development of Academic Attendance Monitoring
establish the ground truth of the minutiae. The System using Fingerprint Identification,” International
minutiae features are detected using FVGFE Journal of Computer Science and Network Security, vol.
algorithm, the Sensitivity (SEN) and Specificity 9, no.5, pp. 164-168, May 2009.
(SPE) are measured over the database. The SEN [7] M. Kass and A. Witkin, “Analysing Oriented Patterns,”
and SPE indicate the ability of the algorithm to Proceedings of Journal of Computer Vision Graphics
Image Process, vol. 37, pp. 362-385, 1987.
detect the true minutiae and reject false minutiae
respectively. [8] Bazen and Gerez, “Extraction of Singular points from
Directional Fields of Fingerprints,” Annual Centre for
Telematics and Information Technology Workshop, vol.
24, pp 905-919, July 2002.
TABLE II. MEAN AND STANDARD DEVIATION
[9] [9] L. Hong, A. K. Jain, S. Pankanti and R. Bolle,
OF SEN AND SPE
“Fingerprint Enhancement,” Proceedings of Third IEEE
FEFD FVGFE Workshop on Applications of Computer Vision, pp. 202-
207, 1996.
SEN SPE SEN SPE
[10] B. G. Sherlock, D. M. Monro and K. Millard, “Fingerprint
Mean 79.8 85.36 80.8 87.29 Enhancement by Directional Fourier Filtering,” IEEE
7.21 12.64 7.98 13.14 Proceedings of Visual Image Signal Process, vol.141, pp.
SD
87-94, April 1994.
[11] E. K. Yun and S. B. Cho, “Adaptive Fingerprint Image
Enhancement with Fingerprint Image Quality Analysis,”

143 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
 (IJCSIS) International Journal of Computer Science and Information Security,
Vol. 6, No. 2, 2009

International conference of Image and Vision Computing, K.B. Raja is an Assistant Professor
pp. 101–110, 2006. in Dept of E&CE University
[12] M. P. Brankica and L. Maˇskovi´c, “Fingerprint Minutiae Visvesvaraya college of Engg,
Filtering Based on Multiscale Directional Information,” Bangalore University, Bangalore.
Proceedings of organized crime: Facts and measures of He obtained his Bachelor of
protection meeting, Police academy Belgrade vol. 20, pp. Engineering and Master of
233-244, August 2007. Engineering in Electronics and
[13] F. A. Afsar, M. Arif and M. Hussain, “Fingerprint Communication Engineering from
Identification and Verification System using Minutiae University Visvesvaraya College of
Matching,” National Conference on Emerging Engineering, Bangalore. He was awarded Ph.D. in Computer
Technologies, pp.141-146, 2004. Science and Engineering from Bangalore University. He has
over 35 research publications in refereed International Journals
[14] G. Jagadeeswar Reddy, T.Jaya Chandra Prasad and M. N. and Conference Proceedings. His research interests include
Giri Prasad, “Fingerprint image denoising using curvelet
Image Processing, Biometrics, VLSI Signal Processing,
transform,” Proceeding of Asian Research Publising
computer networks.
Network Journal of Engineering and Applied Sciences,
vol 3, no 3, pp. 31-35, June 2008
[15] K. Zebbiche and F. Khelifi “Region-Based Watermarking
of Biometrics Images: Case Study in Fingerprint Images,” K R Venugopal is currently the
Proceedings of International Journal of Digital Principal and Dean, Faculty of
Multimedia Broadcasting, pp. 1-13, 2008. Engineering, University
[16] Bhupesh Gour, T. K. Bandopadhyaya, Sudhir Sharma Visvesvaraya College of
“Fingerprint Feature Extraction using Midpoint Ridge Engineering, Bangalore
Contour method and Neural Network”, Proceedings of University, Bangalore. He
International Journal of Computer Science and Network obtained his Bachelor of
Security, vol.8, no.7, pp. 99-103, July 2008. Engineering from University
[17] K. Karu and A.K. Jain, “Fingerprint Classification,” Visvesvaraya College of
Proceedings of Pattern Recognition. vol. 29(3), pp.389- Engineering. He received his Masters degree in Computer
404, 1996. Science and Automation from Indian Institute of Science
[18] Fingerprint Verification Competition, Database Bangalore. He was awarded Ph.D. in Economics from
http://bias.csr.unibo.it/fvc2004. Bangalore University and Ph.D. in Computer Science from
Indian Institute of Technology, Madras. He has a distinguished
[19] L. Hong, Y. Wan and A. Jain, "Fingerprint Image
academic career and has degrees in Electronics, Economics,
Enhancement: Algorithm and Performance Evaluation,”
Law, Business Finance, Public Relations, Communications,
IEEE Pattern Analysis and Machine Intelligence, vol. 20,
Industrial Relations, Computer Science and Journalism. He has
pp. 777-787, 1998.
authored 27 books on Computer Science and Economics, which
include Petrodollar and the World Economy, C Aptitude,
AUTHORS PROFILE Mastering C, Microprocessor Programming, Mastering C++
etc. He has been serving as the Professor and Chairman,
Department of Computer Science and Engineering, University
Lavanya B.N. received her
Visvesvaraya College of Engineering, Bangalore University,
Bachelor of Engineering in
Computer Science and Bangalore. During his three decades of service at UVCE he has
Engineering from Sri over 200 research papers to his credit. His research interests
Jayachamarajendra College of include computer networks, parallel and distributed systems,
Engineering Mysore and Master digital signal processing and data mining.
of Engineering in computer
science from Dr.Ambedkar Institute of Technology
Banglore.She is persuing her Ph.D in Computer Science
and Engineering in JNTU Hyderabad and her area of
interest is biometrics, image processing.

144 http://sites.google.com/site/ijcsis/
ISSN 1947-5500