Research On The Key Technology Of Fingerprint Verification System

In the information age, Biometrics, which refers to identification base on physical features or behavior features, has gained more and more attentions. Fingerprint is an immutable feature and scientists find different finger has different fingerprint, even the twins have different fingerprints. Compared with other identification method such as password, person identification card, fingerprint identification is more reliable and more capable to identify impostor from authority user. This thesis focuses on the research and development of auto fingerprint identification system, especially the preprocessing method of fingerprint images.Firstly the technique background and the future directions of biometrics identification are overviewed in this thesis. Secondly different kinds of biometrics techniques are compared. The technique background and the future directions of fingerprint identification are overviewed. Then we analyzed the preprocessing of fingerprint images, improving the clarity of ridge structures of fingerprint images to facilitate automatic extraction of features. In order to minimize the influence of the noise in the estimation of local ridge orientation, we presented a novel preprocessing methods based on the wavelet domain. Then a method for binarization of the fingerprint images is presented. This method is based on the characteristic of the regions of ridges after the noneffective blocks are excluded in the fingerprint image. The experimental results of this method demonstrate that this algorithm can distinguish useful information from the noise presented in the image and produce accurate binarization results with fewer pseudo-structures.Finally, the feature extraction, finding representative features from an input image for purpose of fingerprint matching, as well as minutiae matching, determining whether two sets of features are extracted from the same finger are discussed in this thesis. All the experiments results are presented and analyzed in this paper. In the end, we proposed some research directions in the future research.