The Arithmetic Study Of Fingerprint Identification

With the accelerating pace of the information society, secure and practical personal identification methods are highly required,information security has becoming a hot research.As biometric technology can not be copied,transplanted and modeled,biometric technology more widely applied in the field of information security.In a variety of biometric,among the numerous biometrics,more attentions have been pain to the fingerprint identification technology due to its convenience,high accuracy and low cost.The technique of fingerprint identification has become one of the widest used biometric identification techniques.It has been widely used in electronic commerce,criminal identification,information safety etc.In this paper, based on the research on many literature of automatic fingerprint identification system, three aspects-fingerprint preprocessing,fingeprrinte feature extraction, fingerprint matching algorithms are deeply researched. The main points of the dissertation are as follows:In the fingerprint preprocessing part, this thesis does deep analysis and comparison on pretreatment's every step, such as normalization, direction calculation, direction filtering,binarization,thinning and so on, having seleeted an appropriate fingerprint preprocessing project.In the feature extraction part, study and realize the algorithm on feature extraction,which is based on traceing the thin ridge expressing the type of current point and the states of its 8-neighborhood pixels by 8-neighbour coding.this algorithm can delete false minutiae and store true minutiae at the same time so that it can improve the efficiency of the algorithm.In fingerprint matching, this thesis uses a matching algorithm based on structure characteristic information.this algorithm utilizes structure characteristic information by characteristic point and its surroundings to build up the characteristic matrix to carry on matching an operation, which has a good robustness to circumrotate, excursion, distort and compress.All the algorithms mentioned above,this thesis has programed and realized them with Matlab7.0 on PC.Emulation experiment get the result of 98.56% in the standard of CR.It indicates that this system can satisfy the basic request of fingerprint identification.These algorithms in this thesis provide a good theories basis and the experiment foundation for further research of developing AFIS.