Hybrid Approach for Robust Fingerprint Recognition by Combining Local Binary Pattern And Principal Component Analysis

Download Full-Text PDF Cite this Publication

Text Only Version

Hybrid Approach for Robust Fingerprint Recognition by Combining Local Binary Pattern And Principal Component Analysis

Dr. Kamini H. Solanki

Parul Institute of Computer Application, Parul University

AbstractFingerprint is used for real time application now days. So reliability is the more important thing for security. Fingerprint is rapidly becoming area of interest in computer science field. In this paper, fingerprint recognition is done using combination of local binary pattern and principal component analysis for better recognition result. The feature extraction is an essential step for image analysis, object representation, visualization, and many other image processing tasks. PCA is used for dimension reduction. LBP is used to measure the expression of Fingerprint. Hybrid approach will increase the recognition rate (RR) of Fingerprint and also decreased verification time and false match rate. So it is most suitable for real time application. We compared proposed method with both PCA and LBP to compute these changes that increased Fingerprint recognition rate and decreased recognition time and false match rate.

Index Terms Fingerprint image representation; LBP; PCA; Recognition rate, False match rate

  1. INTRODUCTION

    Fingerprint Recognition Features can be extracted using global features and local features. Global features focus on the Whole Entire Image so Less Accuracy. Local Features focus on the local features of the Fingerprint, which help to identify and verify the persons using the unique details in the Fingerprint that is more Accurate. The local binary pattern (LBP) [1] was originally designed for texture description. It is invariant to monotonic grey- scale transformations which is very important for texture analysis. Also due to the computational simplicity processing of image in real time is possible. With LBP it is possible to describe the texture and shape of a digital image. This is done by dividing an image into several small regions from which the features are extracted. These features consist of binary patterns that describe the surroundings of pixels in the regions. The obtained features from the regions are concatenated into a single feature histogram, which forms a representation of the image. Images can then be compared by measuring the similarity (distance) between their locally feature extraction and dimension reduction. According to several studies [2-4] Fingerprint recognition using the LBP+PCA method provides very good results, both in terms of speed and recognition performance.

    Fig.1 Fingerprint recognition methods

  2. FINGERPRINT ACQUISITION

    The first challenge facing a finger scan system is to acquire a high quality image of the fingerprint. Image quality is measured by dots per inch (DPI) more DPI means a high resolution image. The lowest DPI generally found is the 300 to 350 DPI range.

  3. FINGERPRINT REPRESENTATION There are seven patterns of papillary ridge i.e. Loop, Arch, Whorl, Tented Arch, Double Loop, Central Pocked Loop and Accidental. From seven patterns of papillary ridge there are three patterns which most common like depicted bellow[19].

    Fig. 2 Few example of papillary ridge pattern

    The human population has fingerprints in the following percentages:

    Loop 65%

    Whorl — 30%

    Arch — 5%

    Fig. 3 Some Fingerprint Patterns

    Fingerprint Basics (minutiae)

    Fig 4. Different minutiae[20]

    Feature Detection for Matching

    Fig. 5 Finger scan minutiae[19]

  4. LOCAL BINARY PATTERNS

    Local Binary Pattern (LBP) operator it is possible to describe the texture and shape of a digital (gray scale) image. One LBP is a binary code for an image-pixel which tells something about the local neighborhood of that pixel [7]. This operator works with the eight neighbors of a pixel, using the value of this center pixel as a threshold. If a neighbor pixel has a higher gray value than the center pixel (or the same gray value) than a one is assigned to that pixel, else it gets a zero. The LBP code for the center pixel is then produced by concatenating the eight ones or zeros to a binary code.

    Fig. 6 The original LBP operator

    Later the LBP operator was extended to use neighborhoods of different sizes. In this case a circle is made with radius R from the center pixel. P sampling points on the edge of this circle are taken and compared with the value of the center pixel. To get the values of all sampling points in the neighborhood for any radius and any number of pixels, for neighborhoods the notation (P, R) is used. Figure 2 illustrates three neighbor-sets for different values of P and

    R. Also it is referred as the multi scale LBP or extended LBP.

    Fig. 7 Circularly neighbor-sets for three different values of P and R.

  5. PCA ALGORITHM

    An Efficient method for Fingerprint recognition is Principal Component Analysis (PCA). The PCA has been extensively employed for Fingerprint recognition algorithms. It is one of the most popular representation methods for a Fingerprint image. It not only reduces the dimensionality of the image, but also retains some of the variations in the image data. The system functions by projecting Fingerprint image onto a feature space that spans the significant variations among known Fingerprint images. The significant features are known as Eigen vector, because they are the eigenvectors (Principal Component) of the set of Fingerprints they do not necessarily correspond to the features. The projection operation characterize an individual Fingerprint by a weighted sum of the Eigen Fingerprints features and so to recognize a particular Fingerprint it is necessary only to compare these weights to those individuals. The Eigen Object Recognizer class applies PCA on each image, the

    results of which will be an array of Eigen values. To perform PCA several steps are undertaken: [8]

    Stage 1: Subtract the Mean of the data from each variable (our adjusted data) subtraction of the overall mean from each of our values as for covariance we need at least two dimensions of data. It is in fact the subtraction of the mean of each row from each element in that row.

    Stage 2: Calculate and form a covariance Matrix

    Stage 3: Calculate Eigenvectors and Eigen values from the covariance Matrix Eigen values are a product of multiplying matrices however they are as special case. Eigen values are found by multiples of the covariance matrix by a vector in two dimensional space (i.e. a Eigenvector). This makes the covariance matrix the equivalent of a transformation matrix.

    Stage 4: Chose a Feature Vector (a fancy name for a matrix of vectors) Once Eigenvectors are found from the covariance matrix, the next step is to order them by Eigen value, highest to lowest. This gives you the components in order of significance. Here the data can be compressed and the weaker vectors are removed producing a lossy compression method, the data lost is deemed to be insignificant.

    Stage 5: Multiply the transposed Feature Vectors by the transposed adjusted data The final stage in PCA is to take the transpose of the feature vector matrix and multiply it with the transposed adjusted data set (the adjusted data set is from Stage 1 where the mean was subtracted from the data).

    PCA on FEI Database

    Fig. 8 PCA Algorithm

  6. COMPARISON OF PCA & LBP ALGORITHM

    PCA Algorithm

    LBP Algorithm

    • High Recognition Rate

    • Global feature extraction

    • Fastest executiontime

    • Suitable for real time application

    • Illumination problem removal

    • Local feature extraction

  7. PROPOSED ALGORITHM

    Step 1: Find the mean image.

    Step 2: Reallocate/instantiate array for the local binary pattern.

    Step 3: LBP feature extraction using Feature Vectors.

    Step 4: Calculate the ordered eigenvectors and eigenvalues

    Step 5: Combine LBP local feature vectors and PCA global feature eigen vectors.

    Step 6: Template generation using these features and stored in database.

    Step 7: Apply Step 1 to Step 5 on Inputted Image.

    Step 8: Local feature verification and Global feature verification applied on Database images and inputted image.

    Step 9 Calculate minimum distances between input image and database images using Ecudian distance measurement method.

    Step 10: Retrieved image from database which have minimum distance between input image and database images.

  8. IMPLEMENTATION OF PROPOSED

    ALGORITHM

    Fig. 9 Implementation of proposed algorithm on real time database

  9. RESULT ANALYSES OF PCA AND LBP

    ALGORITHMS

    Table I. PCA& LBP on Fingerprint Database

    Subje ct

    Databa se Images

    Input Imag es

    Recognitio n Rate (RR%)

    False Match Rate (FMR%)

    AVG.Tim

    e (In

    Seconds)

    PC A

    LB P

    PC A

    LB P

    PC A

    LB P

    05

    14

    14

    98.5

    0

    62.8

    5

    1.5

    37.1

    5

    0.3

    7

    0.0

    5

    10

    14

    14

    95.7

    1

    70.0

    0

    4.29

    30.0

    0

    0.2

    5

    0.0

    5

    15

    14

    14

    92.8

    5

    79.5

    2

    7.15

    20.4

    8

    0.2

    9

    0.0

    5

    20

    14

    14

    89.6

    4

    64.6

    4

    10.3

    6

    35.3

    6

    0.3

    2

    0.0

    5

    Table II. PCA& LBP on Real Time Database

    Subje ct

    Datab ase Images

    Inpu t Imag es

    Recognitio n Rate (RR%)

    False Match Rate (FMR%)

    AVG.Tim

    e (In

    Seconds)

    PC A

    LB P

    PC A

    LB P

    PC A

    LB P

    05

    5

    6

    76.6

    6

    52.2

    2

    23.3

    4

    47.7

    8

    0.2

    2

    0.0

    4

    10

    5

    6

    63.3

    3

    48.6

    6

    36.6

    7

    51.3

    4

    0.2

    5

    0.0

    5

    15

    5

    6

    58.8

    8

    46.6

    6

    41.1

    2

    53.3

    4

    0.2

    7

    0.0

    5

  10. CONCLUSION

LBP is fastest execution operator so it is most suitable for real time application. LBP feature vector used for the removal of illumination problem but it works only on local feature. So it cannot capture dominant features with large- scale structures. PCA has high accuracy rate but it has illumination and pose problem. If LBP feature vector combine with PCA eigen vector, increased recognition rate and decreased false match rate as well as verification time.

REFERENCES

  1. T. Ojala, M. Pietikainen, and D. Harwood, A comparative study of texture measures with classification based on feature distributions, Pattern Recognition, vol. 29 , 1996, pp. 51-59.

  2. T. Ahonen, A. Hadid and M. Pietik¨ainen. Face recognition with Local Binary Patterns. Machine Vision Group, University of Oulu, Finland, 2004.

  3. T. Ahonen, A. Hadid and M. Pietik¨ainen. Face description with Local Binary Patterns: Application to Face Recognition. Machine Vision Group, University of Oulu, Finland, 2006.

  4. T. Ahonen, A. Hadid, M. Pietik¨ainen and T. M¨aenp¨a¨a. Face recognition based on the appearance of local regions. In Proceedings of the 17th International Conference on Pattern Recognition, 2004.

  5. T. Ojala, M. Pjetikainen, Multiresolution Grey-Scale and Rotation Invariant Texture Classification with Local Binary Patterns, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24,pp.971-987, 2002.

  6. T. Ahomen, A. Hadid, and M. Pietikainen, Face recognition with local binary patterns, Computer Vision, ECCV 2004 Proceedings, Lecture Notes in Computer Science 3021, Springer, 2004, pp. 469-481.

  7. T. Ojala, M. Pietik¨ainen and D. Harwood. A comparative study of texture measures with classification based on feature distributions.Pattern Recognition vol. 29, 1996.

  8. Dr. Prashant P. Pittalia ,Mrs. Kamini H. Solanki. An Invention Approach to 3D Fingerprint Recognition using Combination of 2D Texture Data and 3D Shape Data. International Journal of Application or Innovation in Engineering & Management (IJAIEM) Volume 2, Issue 11, November 2013.

  9. Timo Ahonen, Abdenour Hadid, Matti Pietikäinen ,Computer Vision – ECCV 2004 Lecture Notes in Computer Science Volume 3021, 2004, pp 469-481 Fingerprint Recognition with Local Binary Patterns

  10. Shengcai Liao, Xiangxin Zhu, Zhen Lei, Lun Zhang, and Stan Z. Li. Learning Multi-scale Block Local Binary Patterns for Fingerprint Recognition. Center for Biometrics and Security Research & National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences,95 Zhongguancun Donglu, Beijing 100080, China {scliao,xxzhu,zlei,lzhang,szli}

    @nlpr.ia.ac.cn http://www.cbsr.ia.ac.cn

  11. AN EFFICIENT METHOD FOR FINGERPRINT RECOGNITION USING PRINCIPAL COMPONENT ANALYSIS (PCA) Gunjan Dashore, Dr. V.Cyril Ra, International Journal of Advanced Technology & Engineering Research (IJATER), ISSN NO: 2250-3536 VOLUME 2, ISSUE 2, MARCH 2012.

  12. M.-H. Yang, D. Kriegman, and N. Ahuja. Detecting Fingerprints in images:A survey. IEEE Transactions on PatternAnalysis and Machine Intel- ligence, 24(1):3458, January 2002.

  13. Fingerprint Recognition Algorithms Proyecto Fin de Carrera Ion Marqu´es June 16, 2010

  14. A. Jain, R. Duin, and J. Mao. Statistical pattern recognition: A review. IEEE Transactions on Pattern Analysis and Machine Intelli- gence, 22(1):437, January 2000.

  15. PAUL VIOLA , MICHAEL J. JONES ,Real-Time Fingerprint Detection ,International Journal of Computer Vision 57(2), 137 154, 2004 Kluwer Academic Publishers. Manufactured in The Netherlands. Robust viola@microsoft.com, mjones@merl.com

  16. Di Huang, Student Member, IEEE, Caifeng Shan, Member, IEEE, Mohsen Ardabilian, Yunhong Wang, Member, IEEE, and Liming Chen, Member, IEEE Local Binary Patterns and Its Application to Facial Image Analysis: A Survey, IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICSPART C: APPLICATIONS AND REVIEWS, VOL. 41, NO. 6,

    NOVEMBER 2011

  17. Aleix M. MartõÂnez, Member, IEEE, and

  18. Avinash C. Kak, PCA versus LDA, IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 23, NO. 2 FEBRUARY 2001

  19. Overview of Fingerprint Recognition System, Mouad .M.H.Ali Research Scholar at Dr. Babasaheb Ambedkar Marathwada University Aurangabad, (M.S), India, International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT) 2016, 978-1-4673-9939-5/16/$31.00 ©2016 IEEE

  20. Human Identification and Fingerprints: A Review Nitin Kaushal1 * and Purnima Kaushal2, Journal of Biometrics & Biostatistics, Kaushal and Kaushal. J Biomet Biostat 2011, 2:4 DOI: 10.4172/2155-6180.1000123

  21. Fingerprint Recognition System and Tehniques: A Survey, Rahul Sharma, Nidhi Mishra, Sanjeev Kumar Yadav, International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 1670 ISSN 2229-5518, IJSER © 2013 http://www.ijser.org

  22. Initial Results on the Composition of Fingerprints and its Evolution as a Function of Time by GC/MS Analysis, Céline Weyermann Ph.D. ,Claude Roux Ph.D. ,Christophe Champod Ph.D., First published: 12 August 2010, https://doi.org/10.1111/j.1556-4029.2010.01523.x

  23. A fingerprint and finger-vein based cancelable multi-biometric system, Author links open overlay panelYang Wenchenga, Wang Songb, Hu Jiankunc Zheng Guangloua Craig Vallia, Pattern Recognition Volume 78, June 2018, Pages 242-251

Leave a Reply

Your email address will not be published. Required fields are marked *