Research On Recognition Algorithm Of Fingerprint

With the development of electronic computer and information technology, the identification technology using biometric have drawn more attention. Due to the uniqueness and invariability of fingerprints, the recognition based on fingerprints is becoming an attractive alternative to the traditional methods of identification. However, there is much special noise in fingerprint images and many methods wouldn't be publicized, so research on ideal fingerprint recognition system is still an important subject.The algorithm of the preprocessing,feature extraction and feature matching of the fingerprint images are studied in detail in this paper. The formula algorithm used to get the orientation of the fingerprint image, and then the filter with orientation is introduced to filter. In order to reduce the computation of the space and time, the Half Gabor Filter is proposed; the algorithm based on ridge tracing using neighborhood of 8 encoding is introduced to extract minutia. Since the fact that many false minutiae raised after the process of extraction, a post processing algorithm deleting false minutiae using local structural information of points is discussed in detail, also the main false minutiae are deleted; In matching algorithm based on points pattern, a matching algorithm based on bunch distance in this paper is introduced, in order to adapt to nonlinear distortion of the fingerprints, the Bounding Box of variable size is introduced too, and the CR is improved.All algorithms of fingerprint recognition are implemented by C programming on PC. Experiments show that all algorithms perform well and demonstrate that our system meets the given requirements.