![graphical types of ridges and valleys [14]](https://www.researchgate.net/profile/Mouad-Ali/publication/306304227/figure/fig1/AS:614033774354439@1523408468926/graphical-types-of-ridges-and-valleys-14_Q320.jpg)
Content uploaded by Mouad M.H. Ali
Author content
All content in this area was uploaded by Mouad M.H. Ali on Oct 11, 2017
Content may be subject to copyright.
Fingerprint Recognition for Person Identification and
Verification Based on Minutiae Matching
Mouad.M.H.Ali
Research Scholar at Dr.
Babasaheb Ambedkar
Marathwada University
Aurangabad,(M.S), India
Mouad198080@gmail.com
Vivek H. Mahale
Research Scholar at Dr.
Babasaheb Ambedkar
Marathwada University
Aurangabad,(M.S), India
mahalevh@gmail.com
Pravin Yannawar
Department of CSIT Dr.
Babasaheb Ambedkar
Marathwada University
Aurangabad,(M.S),India
pravinyannawar@gmail.com
A. T. Gaikwad
Institute of Management
Studies and Information
Technology, Aurangabad,
(M.S),India
drashokgaikwad@gmail.com
Abstract—There are various types of applications for fingerprint
recognition which is used for different purposes .fingerprint is
one of the challenging pattern Recognition problem. The
Fingerprint Recognition system is divided into four stages. First
is Acquisition stage to capture the fingerprint image ,The second
is Pre-processing stage to enhancement , binarization ,thinning
fingerprint image. The Third stage is Feature Extraction Stage to
extract the feature from the thinning image by use minutiae
extractor methods to extract ridge ending and ridge bifurcation
from thinning.The fourth stage is matching(Identification,
Verification) to match two minutiae points by using minutiae
matcher method in which similarity and distance measure are
used . The algorithm is tested accurately and reliably by using
fingerprint images from different databases. In this paper the
fingerprint databases used are FVC2000 and FVC2002
Databases, we see that ,the FVC2002 database perform better
results compare with FVC2000 database. The recognition system
evaluate with two factor FAR and FRR ,In this system the result
of FAR is 0.0154 and FRR is 0.0137 with Accuracy equal to
98.55%
.
Keywords:-Fingerprint Recognition , Identification ,Verification,
Analysis, Minutiae .
I. Introduction
Fingerprint Recognition is used widely for identification of
the persons as compare to the various biometrics techniques
because of many reasons such as ease of capture, highly
distinctiveness, persistence over time, also the fingerprint
sensors are smaller and cheaper compare with other biometric
sensors .Biometric system is used for person identification by
using his/her characteristics( biological and behavioral).
Biological characteristics are based on the physical part of
human body such as (face ,fingerprint ,iris ,retina and speech)
[2]. The applications such as access control, low enforcement
system ,border management system airport and IT Security
and so on [11]. The behavioral characteristics are based on an
action taken by a human such as (voice, keystroke-scan, and
signature-scan).The defined of fingerprint is a combination of
many of ridges and many of valleys on the fingertip’s
surface[17].in case of the ridge which declare as black lines
and the valleys declare as white lines are shown in Fig 1.
Fig 1. graphical types of ridges and valleys [14]
A fingerprint modality was used many years ago due to
their uniqueness and constancy [16] through a person`s life.
The first step to do fingerprint recognition is enrollment which
is the process to register the biometric data to database as a
template then fingerprint recognition undergo either
Verification process or Identification process which is
depending on the purpose of study. In the verification process
the person’s fingerprint is verified from the database by using
matching algorithms. Also it is called (1:1) Matching. It is the
comparison of a claimant fingerprint against enroll fingerprint,
initially the person enrolls his/her fingerprint into verification
system, and the result show whether the fingerprint which take
from the user is matching with the fingerprint store as a
template in database or not match.
In the case of identification process the fingerprint acquired
from one person is compared with all the fingerprints which
store in database. Also it is called (1:N) matching. it is used in
the process of seeking the criminals. Figure 2 show the process
of enrollment stage, verification stage and identification stage.
The paper consist of seven sections as follow:
The introduction In section I, fingerprint recognition process,
in section II, the proposed system of fingerprint recognition in
section III ,The identification process in section IV, the
verification process in section V, the experimental results are
shown in section VI, finally ,conclusion and future work are
given in section VII.
2016 6th International Advanced Computing Conference
978-1-4673-8286-1/16 $31.00 © 2016 IEEE
DOI 10.1109/IACC.2016.69
324
2016 6th International Advanced Computing Conference
978-1-4673-8286-1/16 $31.00 © 2016 IEEE
DOI 10.1109/IACC.2016.69
324
2016 IEEE 6th International Conference on Advanced Computing
978-1-4673-8286-1/16 $31.00 © 2016 IEEE
DOI 10.1109/IACC.2016.69
324
2016 IEEE 6th International Conference on Advanced Computing
978-1-4673-8286-1/16 $31.00 © 2016 IEEE
DOI 10.1109/IACC.2016.69
332
Fig 2. The main process in enrollment, identification and verification[1].
II. Fingerprint Recognition process
The process is performing of two phases:
a. Training (Enrollment ).
b. Testing(Recognition ).
During the training phase each fingerprint is captured by
biometrics sensor or reader to generate digital image. This
image is used as training data ,then pre-processing apply to
training data for removing unwanted data, noise,
reflection,etc.The pre-processing is used to increase the clarity
of ridge structure.
The output of pre-processing is passed to the Feature
extraction stage for each training data, the feature data can be
extracted and stored in database .During the testing stage the
similar steps as training stage with deference that the features
are compare with the stored features in database to compute
the degree of similarity or score .The most popular method is
minutiae extraction algorithm which we focus in this paper, the
minutiae extraction is a process to find the minutiae in
fingerprint image ,the input fingerprint image in grayscale
image(2-dimensional matrix with range from 0-255). The
minutiae contains (x ,y) coordinate and local ridge orientation
in radian.
There are various types of feature from fingerprint the
minutiae are divided into two types which known as
termination and bifurcation as shows in Fig.1.
In case of termination there is one ridge with two ends
whereas in case of bifurcation the one ridge is divided into two
separate ridges.
III. Proposed System
The fingerprint recognition system is divided into three
stages that are fingerprint image pre-processing, feature
extraction and matching . The matching stage is divide into two
process identification and verification .At the time of capture
fingerprint image the pre-processing stage is applied to it. The
output of this stage will be passed to feature extraction stage
which is extract the minutiae point(ridge ending, Bifurcation)
from thinning fingerprint image, then the false minutiae
removal is applied to extract real minutiae. Finally the real
minutiae is stored in mat lab file (.mat). Then if the fingerprint
is already enroll? then send to matching stage otherwise do
enrolment stage and store it in the database as template. In
identification case(one-to-many matching) , the input feature
set which is matching with N template from database,N
matching will be done. The result will be consider as matching
Score. If matching Score closer to 1 then both fingers from
same user. If matching score near to Zero then both fingers
from deferent user . In verification case(one-to-one matching)
,the input feature set which is matching with one template from
database ,one matching will be done and decided either the
input fingerprint verified or unverified.Figure.3.show the
propose of study .
Fig 3. proposed of study of fingerprint Recognition ,Identification and
Verification
325325325333
The propose system stages are discussed as follow:
A. Acquisition stage
The Acquisition stage is the process to obtain image by
different ways such as Online and Offline .There are number of
methods are used which are discussed in [1].In online method
the optical fingerprint reader is used to capture the image of
fingerprint . In offline method the fingerprint image is obtained
by ink in the area of finger and then put a sheet of white Paper
on fingerprint and scan it to get a digital image. The resolution
of the fingerprint must be 500dpi while the size is(640x480)
pixels. In this paper we have used two standard databases
which are available online(FVC2000 and FVC2002),they
contain 80 fingerprints of 10 different fingers [2,8,9].
B. Pre-processing stage
The Pre-processing stage is the process of removing
unwanted data in fingerprint image such as noise , reflection
,etc. It is used to increase the clarity of ridge structure. The
main steps to do Pre-processing stage are enhancement
fingerprint image , binarization and thinning. The result of
enhancement is shown in the Fig4.For fingerprint
enhancement we applied the following steps :
• Identify ridge segment
• Determine ridge orientations
• Determine ridge frequency
• Apply filters
• Histogram Equalization
• FFT Enhancement [13]
Fig 4. pre-processing stage (a)original image ( b)ridge segment (c)Histogram
Equalization (d) FFT Enhancement
The second step of pre-processing is binarization of
fingerprint image which is a process to transform the image
from 256 levels to two levels(0,1)refers to (black and white)
respectively . The result of binarization is shown in Fig.5. In
this paper we used locally adaptive binarization method which
is summarized in this steps below :
1. The image is divided into blocks with size 16x16.
2. mean intensity value for each block is Calculated.
3. For each pixel the following rule is applied.
Fig 5.The second step of pre-processing,(a) Enhancement image (b)
Binarization.
The third step of pre-processing is thinning which is
shows in Fig.6 . It is also called (skeletonization). To enhance
the binary image the thinning algorithm is used to reduce the
ridges of fingerprint images. There are number of thinning
methods. The most popular thinning algorithms are medial
axis method, contour generation method, local thickness based
thinning approach,sequential and parallel thinning [3 ,4]. We
used morphological operation on binary image ,the main steps
to do thinning is :
1. Clean up the thin image by remove single isolated ,
removes H-Breaks and removes spikes.
2. Remove the connected region at the boundary.
Fig 6. Thinning image
C. Feature Extraction stage
The result of pre-processing stage is passed to the feature
extraction. In this stage feature of image are extracted like
ridges,valleys,minutiae, singular points and etc. These features
are used for verification and identification. The fingerprint
recognition technique is divided to two categories :minutiae
(a) (b)
(c)
(d)
(
a
)
(
b
)
326326326334
based approach [5,6], and pattern-based[7]. In this paper we
used minutiae based approach which consists of two
approaches ,minutiae detection and minutiae matching . There
are various type of minutiae[2].In this paper we use ridge
endings and bifurcation to perform matching approach. The
summaries of algorithm of finding minutiae of fingerprint are
given in the steps as follow :
1) Finding minutiae algorithm
Input: the thinning of fingerprint image, the orientation image
in radians and mask.
Output : Ridge ending, Ridge Bifurcation .
Step1: find the size of thinning image .
Step2:find the label connected components in 2-D binary
image which get the total number of ridge and ridge map.
Step3:scan the thinning fingerprint image to detect the
minutiae ,the 8-neighborhoods pixel are used to determine the
ridge endings and ridge Bifurcation for each block have (0 ,1)
Zero for thinning and one for determine the minutiae.
Step4:if there is one neighbor for the pixel minutiae considered
as ridge ending whereas it is considered as ridge bifurcation if
there are at least 3 neighbors for the pixel.
Step 5: store the ridge endings and ridge Bifurcation in mat lab
file .
Step 6: End .
2) False Minutiae Removal
The false minutiae removal steps are shown in figure below:
ǤǤǤ
After extract the minutiae from fingerprint image by using
minutiae extractor algorithm we passing it to false minutiae
removal algorithm which is given above start with the
centralize the minutiae on window and scan for all the
connected branch of the minutiae and find the false minutiae
structure which are taken from [12]. And the Table I, Show the
false minutiae points in all cases .
Table I .show the false minutiae points
Cas
e
shap Description State
M1
spike pierces into a
valley
Distance between bifurcation and
termination smaller than D (D is
concidered as the average distance
between parallel neighbour
ridges).
M2 a spike falsely
connects two ridges
Two bifurcation present in same
ridge and Distance between them
smaller than D then the both
bifurcation are remove.
M3
Two near
bifurcations present
in the same ridge
M4 Two ridge broken
points separated by
a very short
distance and same
orientation
Distance between two termination
smaller than D and their directions
coincident with samll angel
variation and no other termination
found in between them then they
regarded as fales minutiae and
part of brocken ridge hence
removed.
M5
similar to m4 but
one part of the
broken ridge is so
short .
M6
extension of the m4
and 3
rd
ridge is
found in between
the two parts
M7
a very short ridge
found in the
threshold window
Distance between two termination
of a very sort ridge smaller than
D it is concedered as a false
minutiae and is removed.
D. Matching stage
The matching stage is a process to compare two
fingerprints images(input and template )and compute the
similarity degree between them.
In this paper we use two minutiae set from two fingerprint
image.The matching algorithm is used to know either the two
minutiae set from the same finger or from different finger .
The minutiae matcher based on ridge alignment is used in
this paper in which two images of fingerprint are matched and
a minutiae point from each image is selected to calculate the
similarity of two ridges with them[18]. The system will do the
comparison between the similarity and threshold and when the
similarity more than threshold the new coordination system is
created to which the sets of minutiae point transform. After
apply this step to all minutiae points we get two sets of
transformed minutiae point which are passing finally to
matching algorithm to calculate the matching score by the
following formula:
The matching score is compared with threshold.When the
matching score is greater than threshold the fingerprints are
considered from same person (matching pair)otherwise the
fingerprints are considered from different persons(non
matching pair).In another words if the similarity is near or
equal to one it is matching pair but if the similarity is near to
zero it is non matching pair.
Genuine and Imposter score example:
327327327335
In this case ,we have 10 users with 8 impression ,the total of 80
enrollee attempts .To compute the Genuine and Imposter score
let me to give this simple example with Fig.8 below .
In the case of Genuine score, for one impression user in this
example (8 impression – 1 match= 7 Genuine) and for all
impression for first user is =7 x 8= 56 . now we compute for 10
users the Genuine = 7 x 8 x 10 = 560 Genuine score for all
user in database .
In case of Imposter score , for one impression user in this
example (80 – 8 = 72 Imposter) . for all impression for first
user is =72 x 8= 576 . now we compute for 10 users the
Genuine = 72 x 8 x 10 = 5760 Imposter score for all user in
database . FAR and FRR are calculate using formulas below:
Fig.8. Example of Genuine and Imposter
IV. Identification process
It is the process for comparing between the user of
biometric data and multiple users of template data which take
at enrollment phase. In this process the similarity between
input and all user’s in template database is found. The
Identification process is also known as(1:N)matching. It is
performed when the user provides his/her biometric data and
performed the multiple comparisons from number of user’s to
find the matching. The result will be user’s fingerprint is
identified or not identified.
V. Verification process
It is the process of comparison between the user of
biometric data and one template. The Verification contain
various of biometric data recorded but one of biometric data is
matched. This process also is known as (1:1)matching. The
result will be found or not found.
VI. Experimental Results:
The experiment is performed by using mat lab (R2013a)
and tested on databases FVC 2000 and FVC 2002 [8,9]. The
Table II. show the databases used in our work. figure.10 and
figure.14 show the results of pre-processing stage and feature
extraction stage respectively .The comparison of minutiae
extraction from gray scale image without using enhancement
and with used enhancement is shown in Table III. Fig.11.
shows the matching stage between two fingerprints from same
user and from different users and how much the similarity
score between them. Fig.12. shows GUI of identification (one-
to-many)matching from input and template. The similarity and
distance measure are used to perform the fingerprint
identification, the result is user identified or not.
Fig.13.shows GUI of verification system(one-to-one)matching.
The result will be match or non match. Finally ,the recognition
system shows the result user is recognized or not . To evaluate
the fingerprint recognition system FAR and FRR are
calculated, we used different databases, The first experiment
on FVC2000(DB1_B) which contains 80 images (10 users X 8
impression) ,The second Experiment on FVC2002(DB1_B)
which contains 80 images (10 users X 8 impression. We see
that the FVC 2002 give good result better than FVC2000.The
Table IV. shows the result of FAR,FRR and Accuracy and
Fig.9.show the FAR,FRR and rate of the system. The Table V.
shows the execution time for every stage of the proposes of
study . The formula to calculate Accuracy is shown below:
Table II. Fingerprint Images Databases
Databas
e
Competit
ions
Image
Size
Resolu
tion
Sensor type
DB1_B FVC 2000 300x300 500 dpi Low-cost optical
sensor
DB2_B FVC 2000 256x364 500 dpi Low-cost capacitive
sensor
DB1 _B FVC 2002 388x374 500 dpi Optical sensor
DB2_B FVC 2002 296x560 569 dpi Optical sensor
DB3_B FVC 2002 300x300 500 dpi Capacitiv e sensor
Table III. comparison Minutiae Extraction without /with Enhancement
Images
Minutiae
without Enhancement with Enhancement
Ridge
End
Bifurcation Total Ridge
End
Bifurcation Total
101_1 18 375 393 31 11 42
102_1 10 620 630 70 23 93
103_1 147 343 490 41 30 71
104_1 44 770 814 59 29 88
328328328336
Fig 9. shows the Performance of FAR and FRR
Fig 10.Summary of Pre-processing steps (a)input Image ( b)Histogram of
input (c)FFT Enhancement (d) Histogram Enhancement (e) Binarization (f)
Thinning
(a)
(b)
Fig 11. (a) matching between (101_1) and (101_1) similarity =1.(from
same person).(b)Matching between (101_1) and (102_1) similarity
=0.087121.(from different person).
Fig 12. Fingerprint Identification with similarity score.
Experiments on
Database
Recognition Accuracy
FAR FRR Accuracy
FVC2000 DB1_B 0.2049 0.1944 80.03%
FVC2002 DB1_B 0.0154 0.0137 98.55%
Table IV.The percentage of FAR,FRR and Accuracy
329329329337
(a)
Fig 13. Verification System (1:1) matching
Fig 14. Feature Extraction stage
Table V. Time executing for each stage
Stage
Elapsed time in seconds.
Pre-processing 3.696824
Feature Extraction 0.660177
Matching 0.386971
Identification 4.194027
Verification 0.314468
Recognition 14.885206
Total 24.13767
VII. Conclusion and future work
Our work presented fingerprint identification and
verification based on minutiae features. The work is done in
sequence start from the first stage which is pre-processing
which is used to remove unwanted data and increased the
clarity of ridges of fingerprint image. The second step is the
feature extraction which is used to extract the fingerprint
features. In this work we focus on ridge ending and bifurcation
which is done by using minutiae extractor algorithm .The third
step of this work is the matching which is divided into two
parts identification process also known as(1:N)matching or
verification process also known as(1:1 matching ).Here we
used minutiae matching algorithm with Euclidean distance
measure to find similarity score of two fingerprints images.
The experiments are tested on two fingerprint databases which
are FVC2000 and FVC2002.The result from the experiment 1
on FVC2000 database of FAR and FRR are 0.2049,0.1944
,respectively and result from experiment 2 on FVC2002 of
FAR and FRR are 0.0154,0.0137,respectively.The accuracy
from FVC2000 and FVC2002 are 80.03%,98.55%
respectively. The result of FVC2002 is good comparing with
FVC2000 in this work.
The future work is to do fingerprint identification and
verification by using neural network and fuzzy logic in order to
enhance and evaluate the best performance of fingerprint
recognition system and to create our own database for testing
our work on it .
References
[1] A.K.Jain,A.Ross,and S.Prabhakar,“An introduction to biometric
recognition", IEEE Transaction on Circuits and Systems for Video
Technology, vol. 14, no. 1, pp. 4-20, January 2004.
[2] D. Maltoni, D. Maio, A. K. Jain, S. Prabhakar, Handbook of Fingerprint
Recognition (Second Edition), Springer, London, 2009.
[3] V.Espinosa,“Minutiae Detection Algorithm for Fingerprint
Recognition”, Proc. of XXXV International Carnahan Conference on
Security Technology, pp. 264-266, 16-19 October 2001.
[4] M.Sonka,V.Hlavac,R. Boyle, Image Processing, Analysis, and Machine
Vision, 2
nd
Edition, Pws. Pub. Co, 1998.
[5] J.Ravi,K.B.Raja,K.R.Venugopal,“Fingerprint Recognition Using
Minutia Score Matching”,International Journal of Engineering Science
and Technology, Vol.1 (2), pp. 35-42, 2009.
[6] H. Costin, I. Ciocoiu, T. Barbu, C. Rotariu, “Through Biometric Card in
Romania: Person Identification by Face,Fingerprint and Voice
Recognition”, International Journal of Biomedical Sciences, Volume 1,
Number 4, pp. 264- 269, 2006.
[7] A.Tudosa,M.Costin,T.Barbu,“Fingerprint Recognition using Gabor
filters and Wavelet Features”, Scientific Bulletin of the Politehnic
University of Timisoara, Romania,Transactions on Electronics and
Communications, Tom (49) 63, Fasc. 1, pp. 328-332,2004.
[8] http://bias.csr.unibo.it/fvc2000/.
[9] http://bias.csr.unibo.it/fvc2002/.
[10] S. Mazumdar, V. Dhulipala, "Biometric Security Using Finger Print
Recognition”, University of California, San Diego. 7 pages, Retrieved
30 August 2010.
[11] Gabriel I,Oluwole A, Boniface A, Olatubosun O,” Fingerprint image
enhancement:segmentation to thinning “, Department of Computer
Science. Int J Adv Comp Sci Appl 3(1) 2012 .
[12] Manu Garg and Er. Harish Bansal, “Fingerprint Recognition System using
Minutiae Estimation”, IJAIEM, Volume 2, Issue ISSN 2319– 4847, 5 May
2013.
[13] Amandeep Kaur and Ameeta, Babita,” Minutiae Extraction and Variation of
Fast Fourier Transform on Fingerprint Recognition”,International Journal of
Engineering Research and General Science Volume 2, Issue 6, October-
November, 2014.
[14] Vaibhav Jain and Ajay Kumar Singh, “An approach for Minutia Extraction in
Latent Fingerprint Matching”,International Journal,ISSN:2319–1058,
Volume 6, No. 1 , pp. 51-58, October 2015.
330330330338
[15] Asker M. Bazen,”Fingerprint Identification - Feature Extraction,
Matching, and Database Search”, Ph.D thesis from University of
Twente,
August 19, 2002.
[16] Haiyun Xu, Raymond N.J.Veldhuis, Tom A.M.Kevenaar and Ton
A.H.M.Akkermans,“A Fast Minutiae-Based Fingerprint Recognition
System”,in IEEE system Journal ,Volume 3,No.4,Decmber 2009.
[17] Ashwini R. Patil, Mukesh A. Zaveri, “Novel Approach for Fingerprint
Matching using Minutiae”, in Fourth Asia Int. Conference on
Mathematical/Analytical Modelling and Computer Simulation 2010
IEEE 978-0-7695-4062-7/10. 2010.
[18] Chitresh Saraswat, Amit Kumar, “An Efficient Automatic Attendance
System using Fingerprint Verification Technique”, International Journal
on Computer Science and Engineering(IJCSE), Volume (2) : Issue (2)
pp. 264-269, 2010.
Yannawar Pravin Laxmikan
t
was born in 1979.He
received his B.Sc Computer Science from Dr. Babasaheb
Ambedkar Marathwada University, Aurangabad in June 1999 ,
and M.Sc Computer Science from Department of Computer
Science and Information Technology, Dr. Babasaheb
Ambedkar Marathwada University, Aurangabad in June 2001
and the Ph.D(Computer Science), Department of Computer
Science and Information Technology, Dr. Babasaheb
Ambedkar Marathwada University Aurangabad in March 201.
He is a Member and Editorial of many journals, Paper
Presentation International Conferences(11) , National
Conferences(10), Journal (05) and Interdisciplinary
Research(02) and Workshop Attended(08),recentlyhe is
Assistant Professor, Vision and Intelligent System Lab,
Department of Computer Science and IT, Dr. B.A. M
University, Aurangabad
A.T.Gaikwad was born in 1976. He received The B.Sc degree
from Dr. Babasaheb Ambedkar Marathwada University,
Aurangabad,India ,in 1998 and he received the M.Sc degree
from Dr. Babasaheb Ambedkar Marathwada University,
Aurangabad,India in 2000, MCA degree from Punjab
Technical University ,Aurangabad ,India in 2014 and the Ph.D
degree from Dr. Babasaheb Ambedkar Marathwada University,
Aurangabad,India ,in 2007.He has Published more than ( 27 )
papers in international journals and more than( 08 ) papers in
international conferences .Recently he is a Director of
I.M.S.I.T ,Aurangabad India .
Vivek H. Mahale was born in 1977. He received The B.Sc
degree from Dr. Babasaheb Ambedkar Marathwada
University, Aurangabad,India ,in 1999 and he received the
M.Sc degree from Dr. Babasaheb Ambedkar Marathwada
University, Aurangabad,India in 2002, MCA degree from
Punjab Technical University ,Aurangabad ,India in 2014.He
has Published 10 papers in international journals and 1 papers
in international conferences and 2 paper in national conferenc
. Recently he is a Asst.Professor of I.M.S.I.T ,Aurangabad
India .
Mouad.M.H.Ali was born in1980. He received the B.Sc degree
from Hodeida University Hodeida,Yemen in 2004, and he
received the M.Sc degree from Dr. Babasaheb Ambedkar
Marathwada University, Aurangabad,India in 2013 and Master Of
Multimedia Technology (M.Sc MM) degree from PUNJAB
Technical University, jalandhar, Punjab, India, in 2014.Currently ,
He is a Ph.D Student at Dr.BAM University ,Aurangabad ,India.
331331331339
