International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 2372

ISSN 2229-5518

A Novel Approach To Classification Based

FingerPrint Verification System

Shailendra Patel, Ananya Bhattacharya, Pankaj Kumar

Abstract— Fingerprints are the most oftenly used biometric feature for a person credentials and verification in the field of biometric identification. Fingerprints possess two main types of features that are used for automatic fingerprint identification and verification. First is the Ridge and Furrow structure that forms a special pattern in the central region of the fingerprint and second is the Minutiae details which is associated with the local ridge and furrow structure.In this paper, we have concentrated our implementation on Minutiae based method followed by the discussion about the implementation of a minutiae based matching technique. This approach has been intensively studied, and also is the backbone of the current available fingerprint recognition products. In particular we are interested only in two of the most important minutia features i.e. Ridge Ending and Ridge bifurcation. In a traditional biometric recognition system, the biometric template is usually stored on a central server during enrollment. The candidate biometric template captured by the biometric device is sent to the server where the processing and matching steps are performed. This paper presents an approach to speed up the matching process by classifying the fingerprint pattern into different groups at the time of enrollment, and improves fingerprint matching while matching the input template with stored template. To solve the problem, we take several aspects into consideration like classification of fingerprint, singular points etc. The algorithm result indicates that this approach manages to speed up the matching effectively, and therefore proves to be suitable for large database like forensic divisions.

Index Terms— Biometrics, Identification, Minutiae points, Ridge ending and Bifurcation, Singular points, Verification.

——————————  ——————————

1 INTRODUCTION

ingerprint recognition or fingerprint authentication refers to the automated method of verifying a match between
• Ridge ending - the abrupt end of a ridge
• Ridge bifurcation - a single ridge that divides into 2 ridges

two human fingerprints. Fingerprints are one of many forms of biometrics used to identify an individual and verify their identity. Because of their uniqueness and consistency over time, fingerprints have been used for over a century, more recently becoming automated (i.e. a biometric) due to advancement in computing capabilities. Fingerprint identifica- tion is popular because of the inherent ease in acquisition, the numerous sources (ten fingers) available for collection, and their established use by law enforcement and immigration.

1.1 Fingerprint

A fingerprint is the feature pattern of one finger. It is an impression of the friction ridges and furrows on all parts of a finger. These ridges and furrows present good similarities in each small local window, like parallelism and average width.
However, shown by intensive research on fingerprint recognition, fingerprints are not distinguished by their ridges and furrows, but by features called Minutia, which are some abnormal points on the ridges. Among the variety of minutia types reported in literature, two are mostly significant and in heavy usage:

————————————————

• Shailendra Patel is currently working as a Assistant Professor in GIT, Jaipur, PH +91-9001912315, E-mail:shailu280887@gmail.com

• Ananaya Bhattacharya is currently pursuing masters degree program in Com- puter Sc. And Engg. In Birla Institute of Technology, Mesra Email:

88ananya@gmail.com

• Pankaj Kumar is currently working as a Project Engineer in Infosys. Email:

pankajkumar016@gmail.com

Fig. 1 Minutiae features

1.2 Fingerprint Matching Techniques

The large number of approaches to fingerprint matching can be coarsely classified into three families:

Correlation-based matching: Two fingerprint images are su- perimposed and the correlation between corresponding pixels is computed for different alignments (e.g. various displace- ments and rotations).

Minutiae-based matching: This is the most popular and wide- ly used technique, being the basis of the fingerprint compari- son made by fingerprint examiners. Minutiae are extracted from the two fingerprints and stored as sets of points in the two dimensional plane. Minutiae based matching essentially consists of finding the alignment between the template and the input minutiae sets that results in the maximum number of

International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 2373

ISSN 2229-5518

minutiae pairings.

Pattern-based (or image-based) matching: Pattern based al- gorithms compare the basic fingerprint patterns (arch, whorl, and loop) between a previously stored template and a candi- date fingerprint. This requires that the images be aligned in the same orientation. To do this, the algorithm finds a central point in the fingerprint image and centers on that. In a pat- tern-based algorithm, the template contains the type, size, and orientation of patterns within the aligned fingerprint image. The candidate fingerprint image is graphically compared with the template to determine the degree to which they match.

A typical block diagram of biometric matching systems is shown in Figure below:
pattern types. Henry classification system decreases the effort necessary to search the fingerprint data from a large database. In the Henry fingerprint classification system each finger is assigned a number based on their occurrance in the hand. The right thumb is assigned the number = 1 and it ends with the left little finger (pinky), with its number = 10. In this system each finger is assigned with the numerical values for the finger and this is done for all the fingers in both the hands. Fingers without whorl pattern like arch or loop pattern are assigned the value = 0 (zero) and with whorl are assigned the value = 1. According to the Henry fingerprint classification system, as- signed finger number, finger values and pattern type of finger are shown in Table 1. [5, 6, 7, 11].

TABLE 1

HENRY FINGERPRINT CLASSIFICATION SCHEME

Fig. 2. Block diagram of a typical Fingerprint matching system

In this paper we have discussed about the implementation of a minutia based matching technique. This approach has been intensively studied, also is the backbone of the current available fingerprint recognition products.

2 PREVIOUS WORKS

For fingerprint identification AFIS (Automated Fingerprint Identification System) is used which needs to compare the input fingerprint to the whole fingerprint database. This tech- nique is complex for some applications. In last few years sev- eral scientists have proposed a large number of methods for fingerprint classification and identification. To reduce the number of comparisons and response time, a new scheme has been proposed by scientists Francis Galton and Edward Hen- ry. This scheme is used with existing AFIS database and also used for the systems that require compatibility with human classifications. In this scheme they classified the fingerprint data based on finger ’s features like number of cores and num- ber of deltas. [3, 4]

2.1 Henry Fingerprint Classification system

It permits for logical classification of fingerprints of both hands and fingerprint records grouped based on fingerprint
PGR = [1+(sum of even finger value)]/[1+(sum of odd finger value)]

International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 2374

ISSN 2229-5518

PGR = [1+(2+8)]/1+0 = 11/1
Fingerprint data are stored into database with their relative PGR value. If all the fingers of both hands contain whorl val- ues, then PGR = 31:31, if any person does not contain whorl values then PGR = 1:1. The major drawback of this scheme is that fingerprints of both hands are required. Moreover, be- cause of the storage requirement of fingerprints of both the hands large memory space is required. [9, 10]

3 PROPOSED WORK

The proposed work for fingerprint matching is shown in a block diagram in figure 3. The steps performed for fingerprint enrollment and identification/ verification are shown in the figure below.

3.2 Fingerprint Classification Algorithm

Step1. Find the total number of delta points in fingerprint template. i.e ND.
Step2. If ND=2 the fingerprint is of whorl type.
Step3. Elseif ND=1 and R=1 , the fingerprint is of left loop
type.
Step4. Elseif ND=1 and R=0,the fingerprint is of right loop
type.
Step5. Else fingerprint is of arch type.
Where R is the axis of symmetry.
The flowchart of our proposed algorithm is shown in the fig- ure 5
Fig. 3. Proposed work of typical Fingerprint Matching system

3.1 Fingerprint Classifier

Fingerprint classifiers classify the input fingerprint into four major categories namely Left-Loop, Right-Loop, Whorl and Arch. The proposed classifiers works on the basis of singular point (Delta) extracted. If there are two deltas then it will be counted as whorl or twin loop. If there is no delta then it will be counted as arch. If only one delta is there then it will be either left loop or right loop.
Fig. 5. Flow Chart of Proposed algorithm

We have concentrated our implementation on Minutiae based method. In particular we are interested only in two of the most important minutia features i.e. Ridge Ending and Ridge bifur- cation. (Figure 6)

Fig. 6. a)Ridge Ending (b)Ridge Bifurcation
Fig. 4. Fingerprint Classification

4 RESULTS AND DISCUSSIONS

In implementation, we use MATLAB version 7.10 as an im- plementation platform. This work is mostly based on the Im- age Processing toolbox; it is a part of MATLAB tool and the work is focused on designing a classification based speedup fingerprint identification system that is highly expandable

International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 2375

ISSN 2229-5518

using MATLAB environment. The main benefits of the MATLAB environments are: simplicity, code database, code libraries, PC operation, platform export, User Interface etc. This implementation presents the experimental procedures used for Fingerprint Identification. It also provides the results obtained from the mathematical calculation and discusses the comparison of previous and proposed scheme.

4.1 Experimental Results

A fingerprint database of BIT staff having a total of 1000 fingerprint is used to test the experiment performance. Out of thousand fingerprints 300 are of left loop type,425 are of right loop type,175 are of whorl type and 100 are of arch type. The algorithm is capable of differentiating fingerprints at a good rate by setting an appropriate threshold value.

Fig. 7.(d) Calculation of minutiae points (rigde end and bifurcation) of two fingerprints

Fig. 7.(a) Input fingerprint image and the cassified fingerprint i

Fig. 7.(b) Fingerprint after segmentation,

Fig. 7.(e) Fingerprint after alignment

Fig. 7.(c) Fingerprint image after thinning and skeleton cleaning

Fig. 7.(f) Matching score of two fingerprint

International Journal of Scientific & Engineering Research, Volume 4, Issue 6, June-2013 2376

ISSN 2229-5518

Figure 7(a) shows the input fingerprint image and the cassified fingerprint image. First of all the fingerprint is classified which shows us that to which group the input fingerprint actually belongs. Here a random fingerprint is taken as input, and using the position of singularity points it is computed that the input image belongs to the Right loop class. Then the fingerprint image is segmented from the background by calculating relevant part of fingerprint and contour which is shown in figure 7(b). Figure 7(c) shows fingerprint image after thinning and skeleton cleaning. In this step, morphological skeleton is cleaned and false holes are removed. Then two fingerprints are selected and there minutiae points (rigde end and bifurcation) are calculated as shown in figure 7(d). Figure 7(e) shows the two fingerprints after alignment. Two alignments are performed. In the first alignment, corelation between between skeletons is used to find the best alignment according to the translation. In the second alignment, corelation between skeletons is used to find the best alignment according to the rotation. Figure 7(f ) shows the matching score of two fingerprints by evaluating the distance between the minutiae.
Two indexes are used to determine the performance of a fingerprint verification system: First is FRR (false rejection rate) and the other is FAR (false acceptance rate).
FAR-describes the number of times, someone is inaccurately positively matched, Whereas FRR- derives the number of times, someone who should be identified positively is instead rejected. Table 2 shows the comparative analysis of the two algorithms on the basis of these two indexes.
• FAR
(%) FAR=(FA/N)*100
FA= number of incidents of false acceptance
N=total number of samples
• FRR
(%) FRR=(FR/N)*100
FR=number of incidents of false rejections
N=total number of samples

5 CONCLUSION

In this paper, we proposed a classification algorithm for fingerprints based on ridge pattern and this algorithm leads to proposal of a new scheme of bio-metric identification. Firstly we examined the previous fingerprint classification scheme, and proposed a new identification scheme. After the experimental results it is found that the proposed scheme is faster than previous scheme. It is also concluded that the proposed scheme reduces the memory requirements, because only one hand is used for identification.

ACKNOWLEDGMENT

The authors wish to thank Dr. K. S. Patnaik who is currently working as Assosiate Professor in Birla Institute of Technolo- gy, Mesra in the department of Computer Science and Engi- neering for his valuable guidance during this work.

References

[1] S. M. Metev and V. P. Veiko, Laser Assisted Micro technology, 2nd ed., R. M. Osgood, Jr., Ed. Berlin, Germany: Springer- Verlag, 1998.

[2] J. Breckling, Ed., The Analysis of Directional Time Series: Applications to Wind Speed and Direction, ser. Lecture Notes in Statistics. Berlin, Ger- many: Springer, 1989, vol. 61.

[3] S.C. Dass, Anil K. Jain “Fingerprint classification using orientation field flow Curves” springer,2007,vol 34.

[4] Hanna Choi, S. Lee, D. Moon, Y. Chung, S.Pan “Secret Distribution for Secure Fingerprint Verification” International conference on Convergence and Hybrid Information Technology, IEEE, 2008.

[5] R. Cappelli, D.Maio, D. Maltoni, Loris Nanni “A two stage fingerprint classification system” 2007.

[6] Andrew Senior “A Combination Fingerprint Classifier” IEEE Transac- tions on pattern analysis and machine intelligence, vol. 23, no. 10, OCTO- BER 2001.

[7] “The Henry Classification System” International Biometric Group,

Research Consulting Integration, 2003.

[8] M.T. Leung, W.E. Engeler and P. Frank, “Fingerprint Image Processing Using singularity points”, proc. 10th conf. on Computer and Communication Systems, pp.582-586, Hong Kong 2010.

[9]Dr. H.B. Kekre, Dr. Tanuja Sarode, Rekha Vig “Fingerprint Identification using Sectorized Cepstrum Complex Plane” International Journal of Computer Applications (0975 – 8887) Volume 8– No.1, October

2010.

[10] Mary Lourde R and Dushyant Khosla “Fingerprint Identification in Biometric Security Systems” International Journal of Computer and Electrical Engineering, Vol. 2, No. 5, October, 2010.

[11] Atanu Chatterjee, Shuvankar Mandal, G. M. Atiqur Rahaman, and Abu Shamim Mohammad Arif “Fingerprint Identification and Verification System by Minutiae Extraction Using Artificial Neural Network” JCIT, 2010.