Fingerprint Singularity Detection: A Comparative Study

A singular point or singularity on fingerprint is considered as a fingerprint landmark due its scale, shift, and rotation immutability. It is used for both fingerprint classification and alignment in automatic fingerprint identification systems. This paper presents a comparative study between two singular point detection methods available in the literature. The Poincare index method is the most popular approach, and the complex filter is another proposed method applied on the complex directional images. The maximum complex filter response is highly related to the regions with abrupt changes in the ridge orientations. These regions have a high probability to contain a singular point. The optimum detection method in both processing time and detection accuracy will be updated to suite our efficient classification method. The experimental evaluation for both methods proves that the accuracy achieved by complex filter is up to 95% with considerable processing time compared to 90% with Poincare index method.

[1]  M. M. A. Allah,et al.  A New Fingerprint Classification Approach Based on Fast Fourier Transformer , 2008 .

[2]  Terje Kristensen,et al.  Comparison of neural network based fingerprint classification techniques , 2007, 2007 International Joint Conference on Neural Networks.

[3]  Anil K. Jain,et al.  Handbook of Fingerprint Recognition , 2005, Springer Professional Computing.

[4]  Ali Ismail Awad,et al.  Toward An Efficient Fingerprint Classification , 2011 .

[5]  Dinh Thong Nguyen,et al.  Classification of fingerprints using singular points and their principal axes , 2004, 2004 International Conference on Image Processing, 2004. ICIP '04..

[6]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[7]  Tao Xiong,et al.  A combined SVM and LDA approach for classification , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[8]  Lin Wang,et al.  Application of a new type of singular points in fingerprint classification , 2007, Pattern Recognit. Lett..

[9]  Kai Huang,et al.  Fingerprint Classification Based on Extraction and Analysis of Singularities and Pseudoridges , 2001, VIP.

[10]  Xudong Jiang,et al.  Fingerprint Reference-Point Detection , 2005, EURASIP J. Adv. Signal Process..

[11]  Mark J. T. Smith,et al.  Singular point detection by shape analysis of directional fields in fingerprints , 2006, Pattern Recognit..

[12]  Sankar K. Pal,et al.  Noisy fingerprints classification with directional FFT based features using MLP , 1998, Neural Computing & Applications.

[13]  Manhua Liu,et al.  Fingerprint classification based on Adaboost learning from singularity features , 2010, Pattern Recognit..

[14]  Akio Tojo,et al.  Fingerprint pattern classification , 1984, Pattern Recognit..

[15]  Josef Bigün,et al.  Localization of corresponding points in fingerprints by complex filtering , 2003, Pattern Recognit. Lett..