Fingerprint Authentication Using a Minutiae Score Matching ......Fingerprint, minutiae score...

Peer Reviewed Journal

1

Fingerprint Authentication Using a Minutiae Score Matching Algorithm

JEB STEVEN Q. ABAMONGAhttps://orcid.org/0000-0002-7315-5455

[email protected] of Computer Studies

Saint Michael College of CaragaAtupan St., Nasipit, Agusan del Norte, Philippines

RITCHIE LOU C. COMEhttps://orcid.org/0000-0002-5593-5220



JEAN GOLOSINOhttps://orcid.org/0000-0003-0846-0055



Gunning Fog Index: 10.77 Originality: 99% Grammar Check: 99%Flesch Reading Ease: 50.82 Plagiarism: 1%

ABSTRACT

A biometric factor is something physiologically unique about an individual, such as a fingerprint, facial image, iris, voice pattern, and handwriting. When an individual wants logical or physical access (depending on the implementation), a sample is taken

SMCC Computing JournalISSN Print: 2508-0539 · ISSN Online: 2508-0547

Volume 1 · June 2018

SMCC Computing Journal

2

of the authenticities’ biometric data, for example, a fingerprint. Then, the authenticator, using a previously enrolled version of the same biometric template can match the sample against the stored template to verify the individual’s identity. Biometrics is not secret, as everyone leaves fingerprints everywhere they go, faces and eyes can be photographed, voices can be recorded, and handwriting samples can be obtained. The security of the fingerprint authentication system, therefore, relies on the integrity and authenticity of the biometric information. Therefore careful evaluation must be done for the selection of the fingerprint authentication and Good practices should be followed during the implementation, enrollment, and administration of the fingerprint authentication system. In this paper, we intend to propose a high-speed method for fingerprint recognition based on minutiae matching, which, unlike conventional minutiae matching algorithms, also takes into account region and line structures that exist between minutiae pairs, allowing getting more structural information of the fingerprint and resulting in stronger and more accurate matching of minutiae. For Fingerprint thinning, the Block Filter is used, which scans the image at the boundary to preserve the quality of the image and extract the minutiae from the thinned image. The false matching ratio is better compared to the existing algorithm.

KEYWORDS

Fingerprint, minutiae score matching, authentication system, biometric factor

INTRODUCTION

Identity authentication is becoming more and more important in our society a number of identity authentication techniques have been investigated, including blood vessel patterns in the retina or hand, fingerprint, hand geometry, iris, signature, and voice prints. Among them, fingerprint is one of the most reliable techniques. In this paper, we will introduce an identity authentication system which is capable of authenticating the identity of an individual using his/her fingerprint.

The fingerprint is widely known that a professional fingerprint examiner relies on minute details of ridges structures to match fingerprints. Eighteen different types of local ridges descriptions have been identified. Among them, the two most prominent minutia details that are suitable for fingerprint authentication are ridge endings and ridge bifurcations which are usually called minutiae. Matching fingerprint images is an extremely difficult problem, mainly due to the large variability in different impressions of the same finger. By using minutiae score matching algorithm, our system can identify a person through his/her fingerprint.

The purpose of this study is to authenticate the identity of a person with e same fingerprint using the score matching algorithm, it can identify if the person with the


3

fingerprint is the same person that is registered in the database. The goal of the research is to compare existing algorithms over our proposed algorithm: To design software that has a score matching implementation; to compare the same fingerprint by their minutiae; to design a matching fingerprint using score matching algorithm; to develop a simple user interface for fingerprints that gives accurate output. The study is intended only for Saint Michael College of Caraga for the department criminology. It should not be used for other purposes. The Researcher did not use the hardware device, because the researcher focuses on how to compare the two fingerprints using the Score Matching Algorithm. The Researcher cannot accept fingerprint that captured by the advanced camera, because it can be edited and change the fingerprint

FRAMEWORK

Rao, NagaRaju, Reddy, & Prasad (2008) proposed a fingerprint identification technique using a gray level watershed method to find out the ridges present on a fingerprint image by directly scanned fingerprints or inked impression. Afsar, Arif, & Hussain (2004) developed a method for enhancing the ridge pattern by using a process of oriented diffusion by adaptation of anisotropic diffusion to smooth the image in the direction parallel to the ridge flow. The image intensity varies smoothly as one traverse along the ridges or valleys by removing most of the small irregularities and breaks but with the identity of the individual ridges and valleys preserved. Kocharyan & Sarukhanyan (2001) proposed a method for fingerprint verification which includes both minutiae and model-based orientation field is used. It gives robust discriminatory information other than minutiae points. Fingerprint matching is done by combining the decisions of the matchers based on the orientation field and minutiae.

Wimberly & Liebrock (2011) proposed a method for performance measurement of local operators in fingerprint by detecting the edges of fingerprint images using five local operators namely Sobel, Roberts, Prewitt, Canny and LoG. The edge detected image is further segmented to extract individual segments from the image. Rao, NagaRaju, Reddy, & Prasad (2008) presented a method by introducing a special domain fingerprint enhancement method which decomposes the fingerprint image into a set of filtered images then the orientation field is estimated. A quality mask distinguishes the recoverable and unrecoverable corrupted regions in the input image are generated. Using the estimated orientation field, the input fingerprint image is adaptively enhanced in the recoverable regions.

Kukula, Blomeke, Modi, & Elliott (2009) purposed a method to investigate the effect of five different force levels on fingerprint matching performance, image quality scores, and minutiae count between optical and capacitance fingerprint sensors. Three images were collected from the right index fingers of 75 participants for each sensing technology. Descriptive statistics, analysis of variance, and Kruskal-Wallis nonparametric tests were conducted to assess significant differences in minutiae counts and image


4

quality scores based on the force level. The results reveal a significant difference in image quality score based on the force level and each sensor technology, yet there is no significant difference in minutiae count based on the force levels of the capacitance sensor. The image quality score, shown to be affected by force and sensor type, is one of many factors that influence the system matching performance, yet the removal of low-quality images does not improve the system performance at each force level.

Lomte & Nikam (2013 proposed a method to describe a fingerprint matching based on lines extraction and graph matching principles by adopting a hybrid scheme which consists of a genetic algorithm phase and a local search phase. Experimental results demonstrate the robustness of the algorithm. Ji, Yi, Shang, & Pu, (2007) proposed a method for estimating four direction orientation field by considering four steps, i) preprocessing fingerprint image, ii) determining the primary ridge of fingerprint block using neuron pulse coupled neural network, iii) estimating block direction by projective distance variance of a ridge, instead of a full block, iv) correcting the estimated orientation field. Brar & Singh (2014) used principal graph algorithm to obtain principal curves for auto fingerprint identification system. From principal curves, minutiae extraction algorithm is used to extract the minutiae of the fingerprint. The experimental results show curves obtained from the graph algorithm are smoother than the thinning algorithm. Lumini & Nanni (2008) developed a method for minutiae-based fingerprint and its approach to the problem as two-class pattern recognition. The obtained feature vector by minutiae matching is classified into genuine or imposter by Support Vector Machine resulting remarkable performance improvement Xia, Tong Li, & Zhu (2002) proposed a method to overcome nonlinear distortion using Local Relative Error Descriptor (LRLED). The algorithm consists of three steps i) a pairwise alignment method to achieve fingerprint alignment ii) a matched minutiae pair set is obtained with a threshold to reduce non-matches finally iii) the LRLED – based similarity measure. LRLED is good at distinguishing between corresponding and non-corresponding minutiae-pairs and works well for fingerprint minutiae matching. Ahmed & Ward (2002) presented a method, thinning is the process of reducing the thickness of each line of patterns to just a single pixel width. The requirements of a good algorithm with respect to a fingerprint are i) the thinned fingerprint image obtained should be of single pixel width with no discontinuities ii) Each ridge should be thinned to its central pixel iii) Noise and singular pixels should be eliminated iv) no further removal of pixels should be possible after completion of the thinning process. Ratha, Karu, Chen, & Jain (1996) presented a fingerprint classification system using Fuzzy Neural Network. The fingerprint features such as singular points, positions and direction of core and delta obtained from a binarized fingerprint image. The method is producing good classification results. Rajharia & Gupta (2012) have developed anoid method for Fingerprint recognition. Ridge bifurcations are used as minutiae and ridge bifurcation algorithm with excluding the noise–like points are proposed. Experimental results show the humanoid fingerprint recognition is robust, reliable and rapid.


5

Lie Wei proposed a method for rapid singularities searching algorithm which uses delta field Poincare index and a rapid classification algorithm to classify the fingerprint into 5 classes. The detection algorithm searches the direction field which has the larger direction changes to get the singularities. Singularities detection is used to increase accuracy. George, Abhilash, & Raja (2012) proposed fingerprint enhancement to improve the matching performance and computational efficiency by using an image scale pyramid and directional filtering in the spatial domain. Liu (2010) introduced a structural approach to fingerprint classifications by using the directional image of the fingerprint instead of singularities. The directional image includes a dominant direction of ridge lines.

Raja, (2010) have developed a method for extraction of minutiae from the fingerprint images using midpoint ridge contour representation. The first step is segmentation to separate the foreground from a background of the fingerprint image. A 64 x 64 region is extracted from fingerprint image. The grayscale intensities in 64 x 64 regions are normalized to a constant mean and variance to remove the effects of sensor noise and grayscale variations due to finger pressure differences. After the normalization, the contrast of the ridges is enhanced by filtering 64 x 64 normalized windows by appropriately tuned Gabor filter. The processed fingerprint image is then scanned from top to bottom and left to right and transitions from white (background) to black (foreground) are detected. The length vector is calculated in all the eight directions of contour. Each contour element represents a pixel on the contour, contains fields for the x, y coordinates of the pixel. The proposed method takes less and does not detect any false minutiae.

Ratha, Karu, Chen, & Jain (1996) proposed Scale Invariant Feature Transformation (SIFT) to represent and match the fingerprint. By extracting characteristic SIFT feature points in scale space and perform matching based on the texture information around the feature points. The combination of SIFT and conventional minutiae based system achieves significantly better performance than either of the individual schemes.

Chaudhari, Patnaik, & Patil (2014) have introduced combined methods to build a minutia extractor and a minutia matcher. Segmentation with Morphological operations used to improve thinning, false minutiae removal, minutia marking. Lu, Jiang, & Yau (2002) proposed an effective and efficient algorithm for minutiae extraction to improve the overall performance of an automatic fingerprint identification system because it is very important to preserve true minutiae while removing spurious minutiae in post-processing. The proposed novel fingerprint image post-processing algorithm makes efforts to reliably differentiate spurious minutiae from true ones by making use of ridge number information, referring to the original gray-level image, designing and arranging various processing techniques properly, and also selecting various processing parameters carefully. The proposed post-processing algorithm is effective and efficient.

Jain, Prabhakar, Hong, & Pankanti (2000) have developed a filter-based representation technique for fingerprint identification. The technique exploits both local


6

and global characteristics in a fingerprint to make an identification. Each fingerprint image is filtered in a number of directions and a 640-dimensional feature vector is extracted in the central region of the fingerprint. The feature vector is compact and requires only 640 bytes. The matching stage computes the Euclidian distance between the template finger code and the input finger code. The method gives good matching with high accuracy.

Ballan, Sakarya, & Evans, (1997) introduced Directional Fingerprint Processing using fingerprint smoothing, classification, and identification based on the singular points (delta and core points) obtained from the directional histograms of a fingerprint. Fingerprints are classified into two main categories that are called Lasso and Wirbel. The process includes directional image formation, directional image block representation, singular point detection, and decision. The method gives matching decision vectors with minimum errors, and method is simple and fast.

TECHNICAL BACKGROUND

Image EnhancementTo have an enhancement of the image of the fingerprint has used a local algorithm

of histogram equalization. The image is divided into blocks of pixels of size h x w and for each block is counted the number of pixels at each intensity level (IE histogram) and calculated the new level of intensity for each of the 256 levels, as Equation 1:

Where Bi is the new level of intensity in the block, max intensity is 255, count pixels equal to h w and NJ is a pixel intensity level less than or equal to 1. The image of Figure 2 shows an example of this process.

(a) Original Image. (b) Enhanced Image.

Figure 1. Example of Enhancement of Original Image


7

BinarizationThe pre-processing of Minutiae Score Matching Algorithm uses Binarization to

convert scale image into a binary image by fixing the threshold value. The pixel values above and below the threshold are set to ‘1’ and ‘0’ respectively. An original image and the image after Binarization are shown in the Figure 3.

(a) (b)

Figure 2. (a) Input Fingerprint (b) Binarized Image

Block FilterThe binarized image is thinned using Block Filter to reduce the thickness of all

ridge lines to a single pixel width to extract the minutiae points effectively. Thinning does not change the location and orientation of minutiae points compared to an original fingerprint which ensures accurate estimation of minutiae points. Thinning preserves outermost pixels by placing white pixels at the boundary of the image, as a result, first five and last five rows, first five and last five columns assign the value of one. Dilation and erosion are used to thin the ridges. A binarized Fingerprint and the image after thinning are shown in Figure 3.

(a) (b)

Figure 3. (a) Binarized Fingerprint (b) Image after Thinning


8

ThinningAfter binarization (Figure 4 (a)), is conducting a thinning in the crests of the

image fingerprint so that the minutiae extraction step can better detect endings and bifurcations. Thus, the ridges have 1 pixel of width. The thinning (Figure4 (b)) is accomplished using morphological operations.

(a) Binarization (b) Thinning

Figure 4. Binarization and thinning.

Minutiae ExtractionAfter enhancement, the image of the fingerprint is used for the extraction of

minutiae points. There are several characteristics that can be used to authenticate the fingerprint, but most of the minutiae are restricted to only two types:

The bifurcations and the line end, call terminations. A termination is the point where a line (ridge) ends and junctions are points where the ridge is divided in a simple way to join a ’Y’.

Extraction of all minutiaeThe points of termination and bifurcation are extracted from the thinned image

(Figure 4) of fingerprint with the aid of the concept of Condition Number (Cn). The Equation 5 calculates the condition number for a pixel belonging to the ridge of the fingerprint.

Where,And k represents the eight neighbors of p ordered in a clockwise direction If Cn (p)

is equal to 1, p is a termination point. Since p is a bifurcation, Cn (p) should be equal to 3. All other values of Cn are ignored. At the end have been a two-dimensional array (MINi;j) with the same dimensions I and j of the thinned image containing the values


9

1 and 3 for points minutiae and the remaining values equal to 0. The Figure 5 (a) shows all minutiae extracted from the fingerprint of Figure 4.

Spurious Minutiae FilteringAfter extraction of the minutiae, they pass through a filter to detect spurious

minutiae and remove them. These minutiae are often found in samples of the fingerprint due to the presence of noise in the other stages of image processing and decrease the accuracy and performance of the authentication of the fingerprints. The first part of this filtering is deleting all minutiae belonging to the extremity of fingerprints. To do so creates a binary mask (MB) from the thinned image by performing morphological operations on the same.

Figure 5. Example of spurious minutiae filtering.

Is shown in Figure 5 (b) will be considered true only those minutiae that intercede with the mask. This can be expressed by the following Equation 6:

Min0(i; j) = Min(i; j) xMB(i; j) .

Where Min0(i; j) is the new matrix of minutiae points. The figure 5 (A) shows the result of this operation. The second part of spurious minutiae filtering consists in the analysis of the flow of the ridges, well as the distance and connectivity of the minutiae.

(B) Minutiae image after the spurious minutiae filtering

(A) Mask applied to the array of minutiae points


10

DESIGN AND METHODOLOGY

Figure 6. System Architecture

System Architecture Design this is the actual design of our system. It shows the plan of the system and how it works.

The System Architecture begins with the input of fingerprint image and after that fingerprint image is extracted by the system it includes all operation of image pre-processing: normalization, orientation estimation, edge detection, ridge detection, and thinning and minutiae extraction and it will record into the database.

For enroll, system will add fingerprint information to the database. For the match in dataset, system will select a fingerprint in the database for the user. For update, system will select fingerprint information which user preference, and then allow the user to change its information, but the user cannot change its minutiae information. In comparing the fingerprint the system enrolls the fingerprint into the database and choose the fingerprint that they want to compare then it shows the result.

Activity Diagram

Figure 7. Activity Diagram


11

The activity diagram shows the sequence of activities throughout the data capturing and signal processing modules. The input image is passed to the Data capturing Subsystem for reading image, quality control,and feature extraction. Out of signal processing module is a template created using those extracted features.

The activity diagram describes the behavior of the proposed system in terms of activities. Activities are modeling elements that represent the execution of a set of operations. The execution of an activity can be triggered by the completion of other activities, by the availability of objects, or by external events.

Use case diagram

Figure 8. Use Case Diagram

This use case begins with a user wants to input information and fingerprint image in system. Can load the fingerprint image from an external image file or capture a fingerprint image from a fingerprint scanner.

Initial Authentication ProcessThe user wants to authenticate a fingerprint image. This part includes all operation

of image pre-processing: normalization, orientation estimation, edge detection, ridge detection, and thinning and minutiae extraction. After that user wants to match two fingerprint images. After these two images have pre-processed, system will rotate both of them and then match their minutiae.


12

Database management and Update InformationThe user wants to enroll a fingerprint/match a fingerprint to database/update

fingerprint information/ delete a fingerprint. For enroll, system will add fingerprint information to database. For match in the dataset, system will select a fingerprint in database for user. For an update, the system will select fingerprint information which user preference, and then allow the user to change its information, but the user cannot change its minutiae information. In delete fingerprint information, system will select fingerprint information which user preference, and then allow the user to delete it.

Sequence Diagram

Figure 9. Sequence Diagram

Figure 9 shows the sequence of activities during the initial image storing process. Upon the execution of the software triggered by the user, splash.form will be executed within 3 seconds delay. Upon the termination of the splash. Form, main form will be executed containing the main functions of the software. Each function are triggered through user interaction. The function includes, add.form, view.form, authentication.form and the exit.form.

The process involves comparing the set of minutiae data extracted from the input image to a set of minutiae data extracted from template mage. The matching begins by creating a matrix, called rotate values, of the orientation angle difference between each template minutiae and each input minutiae. The value at rotate values (k,m) represents the difference between the orientation angles of Tk and Im. Tk and Im represent the


13

extracted data in all the columns of row k and row m in the template and input matrices, respectively. Template and input minutiae are selected as reference points for their respective data sets.

Converting to polar coordinates allows for an effective match process conducted regardless of any rotational or translational displacement between the template and input images. The variable rTk represents the radial distance, φT k represents the radial angle, and θTk represents the orientation of the kth minutia, all with respect to the reference minutiae. Also, the variables row Tk and col tk refer to the row and column indices of the kth minutia in the template matrix, while row Tref and col tref refer to the indices of the reference minutia currently being used for the template matrix.

Figure14. Final Result

The figure shows the final result of the matching, after comparing the two fingerprints the system show the result of matching, it gives to the information of the person of the fingerprint that you authenticate.


14

RESULTS AND DISCUSSION

The researcher tested the system with a few fingerprint images the result was given in the table below.

Table 1. Results of testing

Testing (fingerprints) ResultColored Image(captured using digital camera)

4 4

Gray scale Image(acquired using ink)

7 7

Fingerprint extract in the crime scene 4 0

In the implementation of the Researcher. Study shows that score matching algorithm could not compare in the fingerprint in crime scene because the fingerprint in the crime scene is blurred and some of the minutiae have faded because of poor extraction of fingerprint and poor captured by the camera. Example of minutiae fade duringextraction of finger print in the crime scene is the ridge end ridge bifurcation. Figure 15 shows the picture of ridges of fingerprint.

Figure 15 Ridges of Fingerprint

The ridge flowor ridge characteristics revealed in the area of friction ridge detail (mark) are of such low-quantity and/or poor quality that a reliable comparison cannot be made. The area of ridge detail contains insufficient clarity of ridges and characteristics or has been severely compromised by extraneous force.


15

(a) Actual Fingerprint (b) Fingerprint in Crime Scene

Figure 16. Fingerprint image

CONCLUSION

It can be concluded that the use of Minutiae Score Matching Algorithm in stage of matching for fingerprint authentication presented poor results if we compare it in the fingerprint in the crime scene because the fingerprint in the crime scene has blurred image and the original minutiae well missing.

Towards this end, we investigate the problem of maximum matching size ofthe picture and reduce it to the problem of finding the rank of an adjacency matrix, which has the same complexity as that of matrix multiplication. We build minutiae matching score algorithms for rank computation based on Gaussian elimination and Gram-Schmidt process, the complexity of which is cubic in the number vertices in the graph (or the number of minutiae in fingerprints). More advanced techniques of lower asymptotic complexity are also of limited applicability to the problem of fingerprint matching because simpler solutions with cubic complexity outperform them on rather small input sizes (the number of minutiae) used in the fingerprint matching.

RECOMMENDATIONS

Despite the achievements of this project, a number of problems remain to be unsolved in fingerprint recognition. Further research should focus on three specific issues: robust feature extraction from low quality fingerprints, construction of robust templates by the integration of information from multiple prints of the same finger,


16

and issues related to large databases. This section gives an exploration of the possible approaches to this issues. For future work, we recommend implementing this system with a fingerprint hardware device so that more accurate minutiae can record in the database.

LITERATURE CITED

Afsar, F. A., Arif, M., & Hussain, M. (2004, December). Fingerprint identification and verification systemusing minutiae matching. In National Conference on Emerging Technologies (Vol. 2, pp. 141-146). Retrieved on January 16, 2019 from https://goo.gl/X8SMmM

Ahmed, M., & Ward, R. (2002). A rotation invariant rule-based thinning algorithm for character recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(12), 1672-1678. Retrieved on January 18, 2019 from https://goo.gl/cesCD5

Ballan, M., Sakarya, F. A., & Evans, B. L. (1997). A fingerprint classification technique using directional images. In Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on (Vol. 1, pp. 101-104). IEEE. Retrieved on January 18, 2019 from https://goo.gl/3wu77D

Brar, H. S., & Singh, V. P. (2014). Fingerprint Image Recognition Using Bacterial Foraging Optimization Algorithm (BFOA) (Doctoral dissertation). Retrieved on January 18, 2019 from https://goo.gl/kFB1AQ

Chaudhari, A. S., Patnaik, G. K., & Patil, S. S. (2014). Implementation of Minutiae Based Fingerprint Identification System Using Crossing Number Concept. Informatica Economica, 18(1). Retrieved on January 18, 2019 from https://goo.gl/YnH3qd

George, J. P., Abhilash, S. K., & Raja, K. B. (2012). Transform domain fingerprint identification based on DTCWT. Editorial Preface, 3(1). Retrieved on January 18, 2019 from https://goo.gl/ijfzTE

Jain, A. K., Ross, A., & Pankanti, S. (2006). Biometrics: a tool for information security. IEEE transactions on information forensics and security, 1(2), 125-143. Retrieved on January 16, 2019 from https://goo.gl/hLRYhb


17

Jain, A. K., Prabhakar, S., Hong, L., & Pankanti, S. (2000). Filterbank-based fingerprint matching. IEEE transactions on Image Processing, 9(5), 846-859. Retrieved on January 18, 2019 from https://goo.gl/qd4sUj

Ji, L., Yi, Z., Shang, L., & Pu, X. (2007). Binary fingerprint image thinning using template-based PCNNs. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 37(5), 1407-1413. Retrieved on January 18, 2019 from https://goo.gl/ZyMimK

Kocharyan, D., & Sarukhanyan, H. (2001). Feature extraction techniques and minutiae-based fingerprint recognition process. The International Journal of Multimedia Technology, 1(1), 31-35. Retrieved on January 16, 2019 from https://goo.gl/zS24fM

Kukula, E. P., Blomeke, C. R., Modi, S. K., & Elliott, S. J. (2009). Effect of human-biometric sensor interaction on fingerprint matching performance, image quality and minutiae count. International Journal of Computer Applications in Technology, 34(4), 270-277. Retrieved on January 18, 2019 from https://goo.gl/4kmPPb

Liu, M. (2010). Fingerprint classification based on Adaboost learning from singularity features. Pattern Recognition, 43(3), 1062-1070. Retrieved on January 18, 2019 from https://goo.gl/PXLJi6

Lomte, A. C., & Nikam, S. B. (2013). Biometric fingerprint authentication by minutiae extraction using USB token system. International Journal of Computer Technology and Applications, 4(2), 187. Retrieved on January 18, 2019 from https://goo.gl/2Vx9tp

Lumini, A., & Nanni, L. (2008). Advanced methods for two-class pattern recognition problem formulation for minutiae-based fingerprint verification. Pattern Recognition Letters, 29(2), 142-148. Retrieved on January 18, 2019 from https://goo.gl/e6TCBs

Lu, H., Jiang, X., & Yau, W. Y. (2002, December). Effective and efficient fingerprint image postprocessing. In Control, Automation, Robotics and Vision, 2002. ICARCV 2002. 7th International Conference on (Vol. 2, pp. 985-989). IEEE. Retrieved on January 19, 2019 from https://goo.gl/pbJDHq

Paulino, A. A., Feng, J., & Jain, A. K. (2013). Latent fingerprint matching using descriptor-based hough transform. IEEE Transactions on Information Forensics and Security, 8(1), 31-45. Retrieved on January 16, 2019 from https://goo.gl/xKS9hH


18

Raja, K. B. (2010). Fingerprint recognition using minutia score matching. arXiv preprint arXiv:1001.4186. Retrieved on January 18, 2019 from https://goo.gl/GbB8Np

Rao, G. S., NagaRaju, C., Reddy, L. S. S., & Prasad, E. V. (2008). A novel fingerprints identification system based on the edge detection. International Journal of Computer Science and Network Security, 8, 394-397. Retrieved on January 16, 2019 from https://goo.gl/wUwMK9

Ratha, N. K., Karu, K., Chen, S., & Jain, A. K. (1996). A real-time matching system for large fingerprint databases. IEEE Transactions on Pattern Analysis & Machine Intelligence, (8), 799-813. Retrieved on January 18, 2019 from https://goo.gl/teBJLS

Rajharia, J., & Gupta, P. C. (2012). A new and Effective Approach for fingerprint Recognition by using feed forward back propagation neural network. International Journal of Computer Applications, 52(10). Retrieved on January 18, 2019 from https://goo.gl/Kd3qji

Tan, X., Bhanu, B., & Lin, Y. (2005). Fingerprint classification based on learned features. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 35(3), 287-300. Retrieved on January 16, 2019 from https://goo.gl/Pp7KiK

Wimberly, H., & Liebrock, L. M. (2011, May). Using fingerprint authentication to reduce system security: An empirical study. In Security and Privacy (SP), 2011 IEEE Symposium on (pp. 32-46). IEEE. Retrieved on January 16, 2019 from https://goo.gl/DgHzjK

Xia, Y., Tong, H., Li, W. K., & Zhu, L. X. (2002). An adaptive estimation of dimension reduction space. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 64(3), 363-410. Retrieved on January 18, 2019 from https://goo.gl/szN41i

Fingerprint Authentication Using a Minutiae Score Matching ......Fingerprint, minutiae score...

Documents

Transcript of Fingerprint Authentication Using a Minutiae Score Matching ......Fingerprint, minutiae score...