Fingerprint Information Processing And Its Application In Personal Authentication On Networks

With the fast development of electronics,information,computer and networks,biometrics has drawn more and more attention. As one of the most excellent identifying methods in biometrics,fingerprint identification's theory and application appeal a lot of interest. At the present,Internet has spread all over the world. When people enjoy the convenience of freely information exchange and resource share provided by Internet,they have to endure the grave fearful hidden-trouble of information security. Information security has become one of the hottest research topics nowadays,home and abroad. Compared with the traditional passwords,man's biometric features have many advantages,which can help people improve the security of networks. Therefore,how to make use of man's biometric features as personal networks authentication became an important topic home and abroad.Fingerprint information processing is made of fingerprint image's pretreatment,fingerprint matching,fingerprint classification,the storage and communication of fingerprint information and so forth. We have studied pretreatment of fingerprint images with experiments,including all the algorithm implementations of the normalization of fingerprint image,estimation of orientation and frequency field,fingerprint image enhancement,extraction of fingerprint ridges,and thinning of fingerprint ridges. The result can be accepted.During the procedure of fingerprint image processing,the calculation of orientation plays the key role,almost all of the following steps rely on its results. It decides the system's performance. The paper emphasizes the computing algorithm of orientation field. Tests have been carried on traditional methods for orientation computing. The results have shown that the orientation field calculated is not precise,the algorithms not robust,and computing speed not fast. In order to solve this difficulty,this paper provided a new sliding window computing method,which is based on vision and statistic information. Experiments showed this approach can perform calculation more quickly,precisely,and robustly. Not only does this method did well in gray level images,but also in binary images and color images. The results got by this method are satisfying.According to related literature,this paper integrated feature extracting procedure and fingerprint verification algorithm. For the sake of network authentication,data structure of fingerprint feature was defined,and some protocols for network transfer provided. At last,the simulation of fingerprint information transferring was carried. In this course,Winsock was used,and comparative data was transferred.