Study On Image Quality Assessment Method Of Automated Fingerprint Identification System

Biometric authentication is a hotspot among several identity verification method because of its superior security, convenience and easy operation, which includes recognition of fingerprint, voice print, face, iris, palm print and so on. Compared with other biometric authentication technologies, fingerprint recognition is advantageous for its efficiency, convenient collecting and low cost. Particularly, semiconductor fingerprint sense chip emerges in recent years; the fingerprint recognition has begun to be used in embedded equipment.Based on plenty of dissertations and technical reports, this paper provides some researches on the fingerprint image quality assessment system. Firstly, a front-fingerprint image quality measurement is proposed by combining different features of a fingerprint image, e.g. foreground area, wetness, dryness, central position of foreground. Secondly, combining the indices of the number of minutiae, the existence of singular points and front-fingerprint image quality measurement, an integrated fingerprint image quality assessment system is presented. Besides the system of six quality indices, a weighting method is also introduced for finding the final quality value and each quality index. It demonstrates that the scores returned from the integrated system will ensure its reliability to indicate the quality of a given fingerprint image and guarantee the performance of fingerprint recognition system. Singularity detection is based on an improved Poincare index method, and adopts k-means to remove spurious singularity. A new scheme for minutiae detection by following the ridge-line is also presented, where minutiae are extracted directly from the original gray-scale fingerprint images. These methods perform better in terms of efficiency and robustness contrast to traditional algorithms. To evaluate the performance of the proposed approaches, several experiments are conducted on our private system. And the results show effective.At the end of this paper, the problems occurred in actual are given. And problems that are not fully solved were also mentioned.