Conference Paper

Fingerprint Verification Based on Statistical Analysis

Dept. of Comput. Graphics & Multimedia, Univ. Teknol. Malaysia, Skudai, Malaysia
DOI: 10.1109/FUTURETECH.2010.5482771 Conference: Submitted to 5th FutureTech’10
Source: IEEE Xplore

ABSTRACT In this paper, the fingerprint has been analyzed statistically. A sub-image of 129 x 129 was extracted from the original image and transformed into a co-occurrence matrix. Four different type of relative position distances were used to generate the matrices. The results have been analyzed by the Program for Rate Estimation and Statistical Summaries (PRESS). The efficiency of the proposed method has been demonstrated by the experimental results and that the further the distances of the relative position the lower the error equal rate.

Download full-text


Available from: Mohammed Khalil, Sep 27, 2015
36 Reads
  • Source
    • "To measure the minutiae information, the feature-analysis procedure is performed by considering the following two factors: 1) intensity distribution and 2) the relative arrangement of pixels in an image [17]. Here, the crossing number technique is applied to detect the bifurcation and termination points. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Different types of fingerprint detection algorithms that are based on extraction of minutiae points are prevalent in recent literature. In this paper, we propose a new algorithm to locate the virtual core point/centroid of an image. The Euclidean distance between the virtual core point and the minutiae points is taken as a random variable. The mean, variance, skewness, and kurtosis of the random variable are taken as the statistical parameters of the image to observe the similarities or dissimilarities among fingerprints from the same or different persons. Finally, we verified our observations with a moment parameter-based analysis of some previous works.
    Journal of Information Processing Systems 09/2012; vol. 8(no. 3):pp. 421-436. DOI:10.3745/JIPS.2012.8.3.421
  • Source
    • "Classification being achieved using Nearest Cluster Centre classifier with Leave one out method and 3NN classifier. Mohammed Khalil et al., [16] presented statistical analysis of fingerprint images for personal identification. The fingerprint image was enhanced using short time Fourier transform analysis. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Forensic applications like criminal investigations, terrorist identification and National security issues require a strong fingerprint data base and efficient identification system. In this paper we propose DWT based Fingerprint Recognition using Non Minutiae (DWTFR) algorithm. Fingerprint image is decomposed into multi resolution sub bands of LL, LH, HL and HH by applying 3 level DWT. The Dominant local orientation angle {\theta} and Coherence are computed on LL band only. The Centre Area Features and Edge Parameters are determined on each DWT level by considering all four sub bands. The comparison of test fingerprint with database fingerprint is decided based on the Euclidean Distance of all the features. It is observed that the values of FAR, FRR and TSR are improved compared to the existing algorithm.
  • Source
    • "Goats, wolves and lambs are labels commonly applied to problem users. Mohammed S Khalil et al., [22] "
    [Show abstract] [Hide abstract]
    ABSTRACT: The fingerprint is a physiological trait used to identify a person. In this paper, Performance Evaluation of Fingerprint Identification based on DCT and DWT using Multiple Matching Techniques (FDDMM) is proposed. The fingerprint is segmented into four cells of each size 150*240. The DCT is applied on each cell. The Harr Wavelet is applied on DCT coefficient of each cell. The directional information features and centre area features are computed on LL sub band. The final Feature Vector is obtained by concatenating Directional Information and Centre Area Features. The matching techniques viz., ED, SVM, and RF are used to compare test image feature with database image features. It is observed that the values of TSR and FRR are better in the case of proposed algorithm compared to existing algorithm.