The Study On System Of Identifying Fingerprint

Due to their uniqueness and persistence, fingerprints are used as main basis of personal identity. Automated fingerprint identification system(AFIS), a technology of fingerprint identification using computer, is of convenience, high efficiency, security and reliability. It has been applied in many fields such as financial security, data encryption, electronically business and will play a more and more important role in our life.Up till now, affected by the noise and the skin elasticity, fingerprint can not be identified efficiently. The problem how to improve the performance of Automated Fingerprint Identification will be discussed in the paper. In this paper, algorithm will be improved in three aspects : pre-processing; feature extraction; matching.Fingerprint imagine should be normalized before pre-processing so as to provide normative imagine for the following. Forthe sake of directional image of fingerprint, gray variance will be used instead of gray change in the Metre method. An operation about average filtering can make the directional image have more robusness. To get the binarization, we select the valve value by using the maximize entropy.we mainly extract the minutiae features in extracting the fingerprint image features. By computing the value of Poincare Index, we can find the core and delta of the fingerprint. Finally, we can fix on the relative positions of the minutiae according to the core and delta.During the matching, a new method which combine frame information has been used. This method can solve the problem about non-linearity strain of the fingerprint image. It can also raise the system recognition rate.