International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 Fingerprint Feature Extraction Using Hough Transform and Minutiae Extraction Nitika 1, Dr. Nasib Singh Gill 2 1 M. Tech. Student, Department of Computer Science & Application, M.D.U, Rohtak, Haryana (India) nitisuhag@gmail.com 2 Professor, Department of Computer Science & Applications, M.D.U, Rohtak, Haryana (India) nasibsgill@gmail.com Abstract Fingerprint recognition is done with minutiae points and non-minutiae points but, here minutiae points are used for fingerprint recognition. A fingerprint is collection of many ridges and furrows (Valleys). The continuous dark pattern flow in fingerprint is called ridges and the light area between ridges is called furrows. Fingerprint has some unique points on the ridge which is known as minutiae point. The proposed fingerprint identification method is based on the Hough Transform and it enables us to identify a fingerprint even if the scanned image is of poor quality or information about rotation angle is unknown. In this paper we are proposing an algorithm which helps in recognizing of fingerprints by using Hough transform and minutiae extraction. Keywords: Minutiae Extraction, Hough Transforms, Feature Extraction. 1. Introduction Fingerprint recognition or fingerprint authentication refers to the automated method of verifying a match between 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 identification is popular because of the inherent ease in acquisition, the numerous sources (ten fingers) available for collection, and their established use and collections by law enforcement and immigration [4]. A fingerprint-based biometric system is essentially a pattern recognition system that recognizes a person by determining the authenticity of her fingerprint. Depending on the application context, a fingerprintbased biometric system may be called either a verification system or an identification system [3]: 155 A verification system authenticates a person s identity by comparing the captured fingerprints with her own biometric template(s) pre-stored in the system. It conducts one-to-one comparison to determine whether the identity claimed by the individual is true; An identification system recognizes an individual by searching the entire template database for a match. It conducts one-to-many comparisons to establish the identity of the individual. 1.1 What is Fingerprint? Fingerprints are the most important part in biometric for human identification. They are unique and permanent from birth to death. So, fingerprints have been used for the forensic application and personal identification. A fingerprint is collection of many ridges and furrows (Valleys). The continuous dark pattern flow in fingerprint is called ridges and the light area between ridges is called furrows. Fingerprint has some unique points on the ridge which is known as minutiae point [1]. In this paper we can consider two main types of minutiae points which are termination point and bifurcation point as shown in Fig.1. Termination: where a ridge ends and Bifurcation: where ridges split into two parts. Fig. 1 Minutiae Points (Termination, Bifurcation)
International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 1.2 Fingerprint pattern exhibits different types of fingerprint features [6]: a. Level 1 (Global Level): When the ridges are parallel. They are classified as loop, delta, and whorl are shown in Figure 2. Fig. 4 White pores and Sweat pores 2. Fingerprint Recognition Fingerprint whorl pattern Fig. 2 An example of fingerprint image showing core point, delta point and whorl patterns b. Level 2 (Local Level): It is based on minutiae in which the ridges are discontinuous.they are classified as ridge ending, ridge bifurcation, lake, independent ridge, point or island, spur, crossover are shown in fig: crossover are shown in fig: Fingerprint recognition is the process of comparing known fingerprint against another or template fingerprint to determine if the impressions are from the same finger or not. It includes two sub-domains: one is fingerprint verification and the other is fingerprint identification [5]. A verification system authenticates a person s identity by comparing the captured fingerprints with her own biometric template(s) pre-stored in the system. It conducts one-to-one comparison to determine whether the identity claimed by the individual is true; An identification system recognizes an individual by searching the entire template database for a match. It conducts one-to-many comparisons to establish the identity of the individual. 3. Hough Transform Fig. 3An example of fingerprint image showing ending, ridge bifurcation, lake, independent ridge, point or island, spur, crossover. c. Level 3 (Very Fine Level): Intra ridge details are detected [1]. Sweat pores are considered at this level is shown in Figure 4. 156 The Hough transform is a feature extraction technique used in image analysis, computer vision, and digital image processing. The purpose of the technique is to find imperfect instances of objects within a certain class of shapes by a voting procedure. This voting procedure is carried out in a parameter space, from which object candidates are obtained as local maxima in a so-called accumulator space that is explicitly constructed by the algorithm for computing the Hough transform. The classical Hough transform was concerned with the identification of lines in the image, but later the Hough transform has been extended to identifying positions of arbitrary shapes, most commonly circles or ellipses. The Hough transform as it is universally used today was invented by Richard Duda and Peter
International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 Hart in 1972, who called it a "generalized Hough transform"[2] after the related 1962 patent of Paul Hough.[3] The transform was popularized in the computer vision community by Dana H. Ballard through a 1981 journal article titled "Generalizing the Hough transform to detect arbitrary shapes"[8]. 3.1 Fingerprint identification method based on the Hough Transform: The Hough Transform has many positive virtues, which other methods used for image segmentation, do not have. Each point of the image is considered independently. The task of fingerprint identification is made difficult by obstacles, for example unknown rotation angle, missing areas, image defects or displacement. Also a slight scaling problem may occur if the fingerprint pattern was recorded at a young age. The proposed fingerprint identification method is based on the Hough Transform and it enables us to identify a fingerprint even if the scanned image is of poor quality or information about rotation angle is unknown. The assumption of this method is that there is information stored in the database about three different characteristic regions of a fingerprint (3 patterns) and their distances (3 numbers). The identification result is positive if all three patterns and their distances are matched in respect to required threshold. The method has been tested on several images and patterns [9]. 4. Proposed Algorithm Problem definition: Given the test Fingerprint Image the objectives are, 1. Pre-processing the test Fingerprint. 2. Perform the Hough transform 3. Extract the minutiae points. 4. Matching test Fingerprint with the database. Table 1 gives the algorithm for fingerprint verification, in which input test fingerprint image is compared with template fingerprint image, for recognition. Table 1: Algorithm Input: Gray-scale Fingerprint image, template database having r row and c column. Output: Verified fingerprint image with matching score. 1. Input test finger Image say img. 2. Binarized the test image (img). 3. Apply Thinning on Binarized Image(img). 4. Apply Hough transform to get the shape description of img. 5. For i=1:r 6. For j=1:c 7. Binarize the Template image(i,j) then apply thinning 8. Apply Hough transform on template image(i,j) to get shape description 9. If shape description of template image(i,j) matches shape description of img 10. Row=I; 11. Break; 12. End 13. End 14. Extract minutiae point of img say mimg 15. For j=1:c 16. Extract minutiae point of template image(row,j) say t(row,j) 17. If t(row,j)==mimg 18. Then finger matched & exit 19. End 20. If j=c+1 21. Then finger unmatched & exit. 5. Implementation Procedure The above mentioned algorithm is implemented in MATLAB with the details of each step given below. Step1. Input Fingerprint Image The data set consists of fingerprints of 10 persons with 8 alternates of same fingerprint. Firstly, fingerprint acquisition stage is performed. In this, any fingerprint image is taken as input from the dataset. 157
International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 operation, ridges in the fingerprint are converted into black colour while valleys changes into white colour. The Binarisation process involves examining the grey-level value of each pixel in the enhanced image, and if the value is greater than the global threshold, then the pixel value is set to a binary value one otherwise, it is set to zero. Command and description are de scribe s as below: Command: >> im2bw( finger1.jpeg, gray thresh(finger1) Description: This command will convert the image into binary image as shown in figure Fig. 5 input fingerprint image Step 2: Perform Hough Transform Hough transform determine the shape of the template fingerprint image and the test fingerprint image. And then match these both images. The template image which is best match with the test fingerprint image will be taken for further steps. Fig. 7 Binarized image Step 4: Image Thinning Fig. 6 Step 3: Binarize the Resulting Image Binarization is the process to convert gray image into binary image. Most minutiae extraction algorithm works on binary image in which there are only two level of importance i.e. 0 (for black level) and 1 (for white level). After the binarisation The main aim of ridge thinning operation is to eliminate the amount of redundant pixels of ridges until the ridges are having just one pixel wide [11]. Thinning is the morphological process to remove the foreground pixel until they are one pixel wide. So in the first step morphological process apply to reduce the width of the ridge. Two main morphological processes are: Erosion: erosion use to thin object in binary image. Dilation: dilation thins is used to thin the area in valleys in the fingerprint A thinning algorithm is given by [12] in which the thinning operation using two iteration. Each subiteration begins by examining the neighborhood of each pixel in the binary image, and based on a particular set of pixel-deletion criteria, it checks whether the pixel can be deleted or not. These 158
International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 sub-iterations continue until no more can be deleted. Command and description are describes as under: Command:>>bwmorph(binaryfinger1, thin)descripti on: It applies a specific morphological operation ( thin ) to the binary image named binaryfinger1.this command convert this binary image to the thin image. as shown result in figure 8. Where P 9 =P 1. It is defined as half the sum of the differences between pairs of adjacent pixels in the eight-neighborhood. Using the properties of the CN as shown in fig. 12, the ridge pixel can then be classified as a ridge ending, bifurcation or nonminutiae point. For example, a ridge pixel with a CN of one corresponds to a ridge ending, and a CN of three corresponds to a bifurcation [10]. Fig.10 Properties of crossing number Fig. 8 Thinned image Step 5: Extract Minutiae Points from Thinned Image After the fingerprint ridge thinning, the next step is to mark minutia points. The most commonly employed method of minutiae extraction in this category is the Crossing Number (CN) concept. Fig. 9 3X3 neighborhood This method involves the use of the skeleton image where the ridge flow pattern is eight-connected. The minutiae are extracted by scanning the local neighborhood of each ridge pixel in the image using a 3X3 window (fig. 11). The CN value is then computed as follows: =0.5 6. Conclusion Fig. 11 Minutiae Extraction The Hough transform is used to detect the shape of the finger. The purpose of the technique is to find imperfect instances of objects within a certain class of shapes by a voting procedure. Minutiae extraction is used to locate the minutiae points of the fingerprint image. This paper presents the feature extraction using the Hough transform and the minutiae extraction. The most important feature of this method 159
International Journal of Computer Science & Management Studies, Vol. 13, Issue 05, July 2013 is the insensitivity to image interference found in low quality images. In future this technique can be used for fingerprint matching. References [1] Dr. Neeraj Bhargava, Dr. Ritu Bhargava, Manish Mathuria, Pooja DixiT Fingerprint Minutiae Matching using Region of Interest International Journal of Computer Trends and Technology (IJCTT) - volume4issue4 April 2013,ISSN: 2231-2803 http://www.ijcttjournal.org Page 515 [2] Ritu Chhillar MINUTIAE BASED FINGERPRINT RECOGNITION USING FUZZY LOGIC-A REVIEW Volume 4, No. 4, April 2013 Journal of Global Research in Computer Science ISSN-2229-371X. [3] A Tutorial on Fingerprint Recognition1 Davide Maltoni [4] Smital D.Pati and Shailaja A.Patil Fingerprint recognition using minutia matching Proceedings of "Conference on Advances in Communication and Computing (NCACC'12) Held at R.C.Patel Institute of Technology, Shirpur, Dist. Dhule,Maharastra,India. April 21, 2012 [5] Dr. Neeraj Bhargava,Dr. Ritu Bhargava,Prafull Narooka,Minaxi Cotia Fingerprint Recognition Using Minutia Matching International Journal of Computer Trends and Technologyvolume3Issue4-2012 [6] Davide Maltoni, Dario Maio, Anil K. Jain, Salil Prabhakar, Handbook of Fingerprint Recognition, 2nd Edition, Springer Press [7] Krishna Kumar, Basant Kumar, Dharmendra Kumar and Rachna Shah Fingerprint Recognition using Minutiae Extraction [8] D.H. Ballard, "Generalizing the Hough Transform to Detect Arbitrary Shapes", Pattern Recognition, Vol.13, No.2, p.111-122, 1981 [9] Johnathan blackledge, Martin turner Fingerprint and iris identification method Based on the hough transform Biuletyn Instytutu Automatyki I Robotyki Wat Nr 15, 2001 43. [10] Roli Bansal, Priti Sehgal and Punam Bedi Minutiae Extraction from Fingerprint Image- a Review IJCSI International Journal of Computer Science Issues, Vol. 8, Issue 5, No 3, September 2011 ISSN (Online): 1694-0814 www.ijcsi.org. [11] Girish Kulkarni and Dhiraj Patil,Nikhil Patil1 Fingerprint Recognition for Library Management IJCEM International Journal of Computational Engineering & Management, Vol. 16 Issue 1, January 2013 ISSN (Online): 2230-7893 www.ijcem.org IJCEM www.ijcem.org. [12] Pompi Hazarika and David A. Russell, Advances in fingerprinting technology, IEEE Trans. Pattern Anal. Mach. Intell. 12, pp.629 639, 2010 160