AdaptivePoreModelforFingerprintPoreExtraction-JingBao.PDFVIP

Adaptive Pore Model for Fingerprint Pore Extraction ? Qijun Zhao, Lei Zhang, David Zhang, Nan Luo Biometrics Research Centre, Department of Computing, The Hong Kong Polytechnic University, Kowloon, Hong Kong {csqjzhao, cslzhang, csdzhang, csnluo}@comp.polyu.edu.hk Jing Bao Shenzhen Graduate School, Harbin Institute of Technology, China Abstract Sweat pores have been recently employed for auto- mated ?ngerprint recognition, in which the pores are usually extracted by using a computationally expensive skeletonization method or a unitary scale isotropic pore model. In this paper, however, we show that real pores are not always isotropic. To accurately and robustly (a) (b) extract pores, we propose an adaptive anisotropic pore model, whose parameters are adjusted adaptively ac- Figure 1. Partial ?ngerprints of two differ- cording to the ?ngerprint ridge direction and period. ent ?ngers. (a) A closed pore and (b) two The ?ngerprint image is partitioned into blocks and a open pores are marked. local pore model is determined for each block. With the local pore model, a matched ?lter is used to extract the pores within each block. Experiments on a high reso- lution (1200dpi) ?ngerprint dataset are performed and the results demonstrate that the proposed pore model extract and use level-three features in AFRS. In [9], and p