![International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 8 230 – 234
_______________________________________________________________________________________________
230
IJRITCC | August 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
Finger Vein Recognition based on PCA Feature Using Artificial Neural Network
Lovepreet Kaur
Global Institute of Management and Emerging
Technologies, Amritsar
preetlove@gmail.com
Navjot Kaur
Global Institute of Management and Emerging
Technologies, Amritsar
navjot.632@gmail.com
Abstract- Personal recognition technology is developing rapidly as a security system. Traditional methods such as authentication key; password:
card is not secure enough, because they could be stolen or easily forget. Biometrics has been applied to a wide range of systems. According to
various researchers, vein biometrics was a good technique from other biometric authentication system used, such as fingerprints, hand geometry,
voice, etc. of the DNA. Root Authentication systems can be designed in different ways. All methods include the matching stage. A neural
network is an effective way of matching Personal identification authentication system. The finger vein pattern is unique biometric identity of the
human beings. The finger vein recognition is a popular biometric technique which is used for authentication purposes in various applications. In
the propose work an algorithm is proposed to find the accuracy, FRR and FAR of finger vein recognition. The performances of PCA, threshold
segmentation, pre-processing and testing & training techniques has been validate and compared with each other in order to determine the most
accurate results in terms of finger vein recognition.
Keywords: MATLAB, PCA, Threshold based segmentation, ANN and Edge detection Technique.
__________________________________________________*****_________________________________________________
I. INTRODUCTION
Different systems require reliable personal identification
scheme to confirm the identity of the entity or decide to
request him to their services. The reason for this program is
to ensure rendering services only accessed by authorized
users and not others [1]. Examples of such applications
include secure access to building computer systems, laptop
computers, cellular phones and ATMs. In the absence of
strong personal identity program, the system is vulnerable to
the threat of an imposter. Biometric is to to measure the
human physical and behavioral characteristics. The
technology used for identification and access control or
personal identification for monitors. Biometric
authentication basic premise is that each person is unique,
the individual can be identified by its real physical or
behavioral traits. (The term 'biometrics' comes from the
Greek 'bio' means life and 'metric' means measurement). [2].
1.1TYPES OF VARIOUS BIOMETRICS
Biometrics mainly divide into two parts :
1. Physiological
2. Behaviour
Fig. 1 Biometric traits
Behavioral
Keystroke
Signature
Voice
Physiological
Finger print
Hand
Iris
Face
DNA
Biometrics](https://image.slidesharecdn.com/43150468110906-09-2017-180112063614/85/Finger-Vein-Recognition-Based-on-PCA-Feature-using-Artificial-Neural-Network-1-320.jpg)
![International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 8 230 – 234
_______________________________________________________________________________________________
231
IJRITCC | August 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
The physiological characteristic used in this paper is finger
vein along with PCA as a feature extraction technique and
artificial neural network as a classification technique.
1.1.1 Finger vein recognition
Like fingerprints, finger vein based blood vessel patterns are
unique for each individual. Finger vein based blood vessel
pattern have high security because the veins are located
under the surface of the skin. The fingerprints can be
cheated by dummy finger fitted with a copied fingerprint,
but the finger vein based identification system is highly
secure for authentication [3, 4].
1.2 FEATURE EXTRACTION USING PCA
Principle component analysis (PCA) is a statistical process,
to set the value of a collection of observations conversion
using orthogonal transformation related variables called
principal component linear uncorrelated variables. . PCA
can be done by eignvalue of decomposition of a data
covariance (or correlation) matrix characteristic value
decomposition or singular value decomposition of the data
matrix to complete, usually in the middle of the averaging of
the data matrix for each character (and normalized or Z
score) . PCA defines a new orthogonal coordinate system,
which is best described by single data set differences [5].
1.3 ANN in vein recognition
Artificial neural networks are used to identify the vein,
because of their simplicity. Vein identification method can
be obtained after a training match mode. The neural network
is an intelligent system that provides to produce output on
the basis of their training data in the form of vein [6,7].
II. RELATED WORK
S.M.Rajbhoj, and P.B.Mane, [8] proposed a novel multi-
biometric system using two most used biometric traits
fingerprint along with iris. Feature vector of every trait has
been used removed from texture pattern of biometric
images, using DWT and PCA method. Classification of
these feature vectors is carried out using Euclidean distance.
Jinfeng Yang ,Yihua Shi et al.[9], propose a novel Vein
zone enhanced program and finger vein Network
segmentation by using scattering method is intended
Removal, vein directional filtering and misinformation to
effectively enhance finger vein Image. Sepehr
Damavandinejadmonfared et al.[10] propose a simple
algorithm for Finger vein recognition. Principal Component
analysis (PCA), the main part of the kernel Analysis
(KPCA) and kernel entropy component analysis (KECA)
performances have been evaluated. Based on these results,
KPCA has been matched and has been proved that KPCA
method performed well for identifying finger vein. Pedro
Tome, et al. [11] proposed an Open source finger vein
framework. This system was vulnerable to the spoofing
attacks having false acceptance rate is 86%. Lu Yang et al.
[12] proposed a finger vein recognition system used for
identifying the internal characteristics of the living body.
Masaki Watanabe et al. [13] proposed a palm vein
recognition system. Palm scanned three times during the
registration, and then only one final scan is permitted to
confirm authentication. This method is accurate and secure
due to unique blood vessel pattern
III. PROPOSED WORK
In the propose work, vein recognition system using (PCA)
principle component analysis with Artificial Neural Network
(ANN) is presents. The parameters like FAR, FRR and
Accuracy have been calculated.
False Acceptance rate (FAR)
(Total Number of Samples
− Number of Samples that falsely accepted)
/(Total number of samples)
False Rejection Rate (FRR)
(TotalNumber of Samples
− Number of Samples that falsely rejetced)
/(Total number of samples)
Accuracy
100 − (𝐹𝐴𝑅 + 𝐹𝑅𝑅) %
1.1 METHODOLOGY
Thus methodology includes various steps for the
fingervein recognition using MATLAB, which are
listed below:
Step 1: Develop and design a particular GUI for proposed
vein recognition system
Step 2: Upload vein images for training and testing
Step 3: Apply pre-processing step.
Step 4: A code is developed for vein segmentation and
localization. For vein recognition segmentation threshold
based segmentation is used.](https://image.slidesharecdn.com/43150468110906-09-2017-180112063614/85/Finger-Vein-Recognition-Based-on-PCA-Feature-using-Artificial-Neural-Network-2-320.jpg)
![International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 8 230 – 234
_______________________________________________________________________________________________
233
IJRITCC | August 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
Table 1: Results of proposed work
Image No. FAR FRR ACCURACY
1 6.75018 0.789046 92.460436
2 6.67017 0.778035 92.34671
3 5.67429 0.631904 93.21654
4 6.78992 0.660562 93.02471
5 7.45282 0.821064 94.21430
6 6.74522 0.882146 92.32561
7 5.67872 0.594201 94.15628
8 5.34310 0.557211 94.63876
9 6.76384 0.627653 93.76521
10 7.45379 0.792501 92.59401
Fig.4 comparison bar graph between FAR and FRR
The above figure shows the graph between FAR/FRR and
their values. In the figure 5.16 red lines indicates FRR
results and blue lines indicate FAR result. The average value
obtained for FAR is 6.532205 and the average value for
FRR is 0.713432.
Fig.5 Graphical representation of Accuracy
Graphical representation of accuracy obtained for the
proposed work is shown in the figure 5.17.The average
value of the accuracy obtained for the proposed work is
93.27426.
V. CONCLUSION
In the Proposed work finger vein recognition is done by
using PCA, threshold based segmentation using Artificial
neural network and Edge detection technique. In the finger
vein recognition biometric parameters has been calculated
like FAR, FRR and Accuracy using Human finger Vein
pattern under the skin’s surface. Finger vein recognition is
used to identify the individual person and their identity.
PCA technique is used to find principle feature of the
uploaded vein image. Training panel will do one time
simulation whereas testing panel will do it multiple times.
Training and Testing is done for matching the biometric
parameters like if parameters are matched the image get
recognized otherwise not-recognized. Pre-processing is done
for resizing and for color conversion of the loaded image.
The average value of FAR is 6.532205, FRR is 0.713432
and for accuracy is 93.27426.In future we can use
optimization technique for feature optimization. We can use
like generic algorithm, PSO, ACO and BCO. Feature
optimization gives us best optimal feature set. It gives best
fitness results than the PCA which i have applied.
References
[1] Tome, Pedro, Matthias Vanoni, and Sébastien
Marcel,”On the vulnerability of finger vein recognition to
spoofing,”Biometrics Special Interest Group (BIOSIG),
2014 International Conference of the. IEEE, 2014
[2] He Zheng, Qiantong Xu, Yapeng Ye and Wenxin Li,
“Effects of meteorological factors on finger vein
recognition,” 2017 IEEE International Conference on
Identity, Security and Behavior Analysis (ISBA), New
Delhi, 2017, pp. 1-8.
[3] L. Yang; G. Yang; Y. Yin; X. Xi, “Finger Vein
Recognition with Anatomy Structure Analysis,” in IEEE
Transactions on Circuits and Systems for Video
Technology , vol.3 PP.99-105.
[4] M. Sapkale and S. M. Rajbhoj, “A biometric
authentication system based on finger vein
recognition,” 2016 International Conference on Inventive
Computation Technologies (ICICT), Coimbatore, 2016,
pp. 1-4.
[5] T. S. Beng and B. A. Rosdi, “Finger-vein identification
using pattern map and principal component
analysis,” 2011 IEEE International Conference on Signal
and Image Processing Applications (ICSIPA), Kuala
Lumpur, 2011, pp. 530-534.
[6] Wu, Jian-Da, and Siou-Huan Ye. “Driver identification
using finger-vein patterns with Radon transform and
neural network.” Expert Systems with
Applications Vol.36, 2009,pp. 5793-5799.
[7] Wu, Jian-Da, and Chiung-Tsiung Liu. “Finger-vein
pattern identification using principal component analysis](https://image.slidesharecdn.com/43150468110906-09-2017-180112063614/85/Finger-Vein-Recognition-Based-on-PCA-Feature-using-Artificial-Neural-Network-4-320.jpg)
![International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169
Volume: 5 Issue: 8 230 – 234
_______________________________________________________________________________________________
234
IJRITCC | August 2017, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
and the neural network technique.” Expert Systems with
Applications Vol. 38, 2011,pp. 5423-5427.
[8] S.M.Rajbhoj, and P.B.Mane, “Match score integration of
iris and fingerprint in multi-biometrics system,” in
Electronics and Communication Systems (ICECS), 2014
International Conference on, vol.5, 2014, pp.1-5.
[9] Yang, Jinfeng, and Yihua Shi,”Finger-vein network
enhancement and segmentation,” Pattern Analysis and
Applications Vol.17, 2014, pp. 783-797.
[10] Damavandinejadmonfared, Sepehr, et al,”Finger vein
recognition using PCA-based methods,” World Academy
of Science, Engineering and Technology Vol. 6, 2012,
pp. 1079-1081.
[11] Tome, Pedro, Matthias Vanoni, and Sébastien
Marcel,”On the vulnerability of finger vein recognition to
spoofing,”Biometrics Special Interest Group (BIOSIG),
2014 International Conference of the. IEEE, 2014.
[12] Yang, Jinfeng, and Xu Li,”Efficient finger vein
localization and recognition,” Pattern Recognition
(ICPR), 2010 20th International Conference on. IEEE,
2010.
[13] Watanabe, Masaki, et al,”Palm vein authentication
technology and its applications,” Proceedings of the
biometric consortium conference. 2005.](https://image.slidesharecdn.com/43150468110906-09-2017-180112063614/85/Finger-Vein-Recognition-Based-on-PCA-Feature-using-Artificial-Neural-Network-5-320.jpg)
Finger Vein Recognition Based on PCA Feature using Artificial Neural Network
- 1. International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 5 Issue: 8 230 – 234 _______________________________________________________________________________________________ 230 IJRITCC | August 2017, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Finger Vein Recognition based on PCA Feature Using Artificial Neural Network Lovepreet Kaur Global Institute of Management and Emerging Technologies, Amritsar preetlove@gmail.com Navjot Kaur Global Institute of Management and Emerging Technologies, Amritsar navjot.632@gmail.com Abstract- Personal recognition technology is developing rapidly as a security system. Traditional methods such as authentication key; password: card is not secure enough, because they could be stolen or easily forget. Biometrics has been applied to a wide range of systems. According to various researchers, vein biometrics was a good technique from other biometric authentication system used, such as fingerprints, hand geometry, voice, etc. of the DNA. Root Authentication systems can be designed in different ways. All methods include the matching stage. A neural network is an effective way of matching Personal identification authentication system. The finger vein pattern is unique biometric identity of the human beings. The finger vein recognition is a popular biometric technique which is used for authentication purposes in various applications. In the propose work an algorithm is proposed to find the accuracy, FRR and FAR of finger vein recognition. The performances of PCA, threshold segmentation, pre-processing and testing & training techniques has been validate and compared with each other in order to determine the most accurate results in terms of finger vein recognition. Keywords: MATLAB, PCA, Threshold based segmentation, ANN and Edge detection Technique. __________________________________________________*****_________________________________________________ I. INTRODUCTION Different systems require reliable personal identification scheme to confirm the identity of the entity or decide to request him to their services. The reason for this program is to ensure rendering services only accessed by authorized users and not others [1]. Examples of such applications include secure access to building computer systems, laptop computers, cellular phones and ATMs. In the absence of strong personal identity program, the system is vulnerable to the threat of an imposter. Biometric is to to measure the human physical and behavioral characteristics. The technology used for identification and access control or personal identification for monitors. Biometric authentication basic premise is that each person is unique, the individual can be identified by its real physical or behavioral traits. (The term 'biometrics' comes from the Greek 'bio' means life and 'metric' means measurement). [2]. 1.1TYPES OF VARIOUS BIOMETRICS Biometrics mainly divide into two parts : 1. Physiological 2. Behaviour Fig. 1 Biometric traits Behavioral Keystroke Signature Voice Physiological Finger print Hand Iris Face DNA Biometrics
- 2. International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 5 Issue: 8 230 – 234 _______________________________________________________________________________________________ 231 IJRITCC | August 2017, Available @ http://www.ijritcc.org _______________________________________________________________________________________ The physiological characteristic used in this paper is finger vein along with PCA as a feature extraction technique and artificial neural network as a classification technique. 1.1.1 Finger vein recognition Like fingerprints, finger vein based blood vessel patterns are unique for each individual. Finger vein based blood vessel pattern have high security because the veins are located under the surface of the skin. The fingerprints can be cheated by dummy finger fitted with a copied fingerprint, but the finger vein based identification system is highly secure for authentication [3, 4]. 1.2 FEATURE EXTRACTION USING PCA Principle component analysis (PCA) is a statistical process, to set the value of a collection of observations conversion using orthogonal transformation related variables called principal component linear uncorrelated variables. . PCA can be done by eignvalue of decomposition of a data covariance (or correlation) matrix characteristic value decomposition or singular value decomposition of the data matrix to complete, usually in the middle of the averaging of the data matrix for each character (and normalized or Z score) . PCA defines a new orthogonal coordinate system, which is best described by single data set differences [5]. 1.3 ANN in vein recognition Artificial neural networks are used to identify the vein, because of their simplicity. Vein identification method can be obtained after a training match mode. The neural network is an intelligent system that provides to produce output on the basis of their training data in the form of vein [6,7]. II. RELATED WORK S.M.Rajbhoj, and P.B.Mane, [8] proposed a novel multi- biometric system using two most used biometric traits fingerprint along with iris. Feature vector of every trait has been used removed from texture pattern of biometric images, using DWT and PCA method. Classification of these feature vectors is carried out using Euclidean distance. Jinfeng Yang ,Yihua Shi et al.[9], propose a novel Vein zone enhanced program and finger vein Network segmentation by using scattering method is intended Removal, vein directional filtering and misinformation to effectively enhance finger vein Image. Sepehr Damavandinejadmonfared et al.[10] propose a simple algorithm for Finger vein recognition. Principal Component analysis (PCA), the main part of the kernel Analysis (KPCA) and kernel entropy component analysis (KECA) performances have been evaluated. Based on these results, KPCA has been matched and has been proved that KPCA method performed well for identifying finger vein. Pedro Tome, et al. [11] proposed an Open source finger vein framework. This system was vulnerable to the spoofing attacks having false acceptance rate is 86%. Lu Yang et al. [12] proposed a finger vein recognition system used for identifying the internal characteristics of the living body. Masaki Watanabe et al. [13] proposed a palm vein recognition system. Palm scanned three times during the registration, and then only one final scan is permitted to confirm authentication. This method is accurate and secure due to unique blood vessel pattern III. PROPOSED WORK In the propose work, vein recognition system using (PCA) principle component analysis with Artificial Neural Network (ANN) is presents. The parameters like FAR, FRR and Accuracy have been calculated. False Acceptance rate (FAR) (Total Number of Samples − Number of Samples that falsely accepted) /(Total number of samples) False Rejection Rate (FRR) (TotalNumber of Samples − Number of Samples that falsely rejetced) /(Total number of samples) Accuracy 100 − (𝐹𝐴𝑅 + 𝐹𝑅𝑅) % 1.1 METHODOLOGY Thus methodology includes various steps for the fingervein recognition using MATLAB, which are listed below: Step 1: Develop and design a particular GUI for proposed vein recognition system Step 2: Upload vein images for training and testing Step 3: Apply pre-processing step. Step 4: A code is developed for vein segmentation and localization. For vein recognition segmentation threshold based segmentation is used.
- 3. International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 5 Issue: 8 230 – 234 _______________________________________________________________________________________________ 232 IJRITCC | August 2017, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Step 5: After this code is developed using PCA principle component analysis for extracting feature .Now the proposed classifier with appropriate feature sets get trained. Step 6: Initialized Artificial neural network with feature sets of segmented vein region of finger image as an input to artificial neural network. Step 7: After the training of vein we can simulate the proposed work with the test image and check their performance metrics. Step 8: After the classification of testing vein the performance metrics like FAR, FRR and Accuracy are computed. Fig. 2 Flowchart of proposed work IV. SIMULATION RESULTS The simulation has been carried out in an Intel Core 2 Duo CPU system with 2.10 GHz on a 32-bit Windows 7 Ultimate Operating System using MATLAB.CASIA Multi-Spectral finger vein image database contains different types images captured from different people using a self-designed multiple spectral imaging device for the research purpose. All finger vein images are 8 bit gray-level JPEG files without any compression. The parameters that have been calculated are given in the tabular form: Fig. 3 Different stages of vein recognition Start Pre-Processing Feature extraction Using PCA Initialize ANN Training Testing Matching End Calculating parameters Data base
- 4. International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 5 Issue: 8 230 – 234 _______________________________________________________________________________________________ 233 IJRITCC | August 2017, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Table 1: Results of proposed work Image No. FAR FRR ACCURACY 1 6.75018 0.789046 92.460436 2 6.67017 0.778035 92.34671 3 5.67429 0.631904 93.21654 4 6.78992 0.660562 93.02471 5 7.45282 0.821064 94.21430 6 6.74522 0.882146 92.32561 7 5.67872 0.594201 94.15628 8 5.34310 0.557211 94.63876 9 6.76384 0.627653 93.76521 10 7.45379 0.792501 92.59401 Fig.4 comparison bar graph between FAR and FRR The above figure shows the graph between FAR/FRR and their values. In the figure 5.16 red lines indicates FRR results and blue lines indicate FAR result. The average value obtained for FAR is 6.532205 and the average value for FRR is 0.713432. Fig.5 Graphical representation of Accuracy Graphical representation of accuracy obtained for the proposed work is shown in the figure 5.17.The average value of the accuracy obtained for the proposed work is 93.27426. V. CONCLUSION In the Proposed work finger vein recognition is done by using PCA, threshold based segmentation using Artificial neural network and Edge detection technique. In the finger vein recognition biometric parameters has been calculated like FAR, FRR and Accuracy using Human finger Vein pattern under the skin’s surface. Finger vein recognition is used to identify the individual person and their identity. PCA technique is used to find principle feature of the uploaded vein image. Training panel will do one time simulation whereas testing panel will do it multiple times. Training and Testing is done for matching the biometric parameters like if parameters are matched the image get recognized otherwise not-recognized. Pre-processing is done for resizing and for color conversion of the loaded image. The average value of FAR is 6.532205, FRR is 0.713432 and for accuracy is 93.27426.In future we can use optimization technique for feature optimization. We can use like generic algorithm, PSO, ACO and BCO. Feature optimization gives us best optimal feature set. It gives best fitness results than the PCA which i have applied. References [1] Tome, Pedro, Matthias Vanoni, and Sébastien Marcel,”On the vulnerability of finger vein recognition to spoofing,”Biometrics Special Interest Group (BIOSIG), 2014 International Conference of the. IEEE, 2014 [2] He Zheng, Qiantong Xu, Yapeng Ye and Wenxin Li, “Effects of meteorological factors on finger vein recognition,” 2017 IEEE International Conference on Identity, Security and Behavior Analysis (ISBA), New Delhi, 2017, pp. 1-8. [3] L. Yang; G. Yang; Y. Yin; X. Xi, “Finger Vein Recognition with Anatomy Structure Analysis,” in IEEE Transactions on Circuits and Systems for Video Technology , vol.3 PP.99-105. [4] M. Sapkale and S. M. Rajbhoj, “A biometric authentication system based on finger vein recognition,” 2016 International Conference on Inventive Computation Technologies (ICICT), Coimbatore, 2016, pp. 1-4. [5] T. S. Beng and B. A. Rosdi, “Finger-vein identification using pattern map and principal component analysis,” 2011 IEEE International Conference on Signal and Image Processing Applications (ICSIPA), Kuala Lumpur, 2011, pp. 530-534. [6] Wu, Jian-Da, and Siou-Huan Ye. “Driver identification using finger-vein patterns with Radon transform and neural network.” Expert Systems with Applications Vol.36, 2009,pp. 5793-5799. [7] Wu, Jian-Da, and Chiung-Tsiung Liu. “Finger-vein pattern identification using principal component analysis
- 5. International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 5 Issue: 8 230 – 234 _______________________________________________________________________________________________ 234 IJRITCC | August 2017, Available @ http://www.ijritcc.org _______________________________________________________________________________________ and the neural network technique.” Expert Systems with Applications Vol. 38, 2011,pp. 5423-5427. [8] S.M.Rajbhoj, and P.B.Mane, “Match score integration of iris and fingerprint in multi-biometrics system,” in Electronics and Communication Systems (ICECS), 2014 International Conference on, vol.5, 2014, pp.1-5. [9] Yang, Jinfeng, and Yihua Shi,”Finger-vein network enhancement and segmentation,” Pattern Analysis and Applications Vol.17, 2014, pp. 783-797. [10] Damavandinejadmonfared, Sepehr, et al,”Finger vein recognition using PCA-based methods,” World Academy of Science, Engineering and Technology Vol. 6, 2012, pp. 1079-1081. [11] Tome, Pedro, Matthias Vanoni, and Sébastien Marcel,”On the vulnerability of finger vein recognition to spoofing,”Biometrics Special Interest Group (BIOSIG), 2014 International Conference of the. IEEE, 2014. [12] Yang, Jinfeng, and Xu Li,”Efficient finger vein localization and recognition,” Pattern Recognition (ICPR), 2010 20th International Conference on. IEEE, 2010. [13] Watanabe, Masaki, et al,”Palm vein authentication technology and its applications,” Proceedings of the biometric consortium conference. 2005.