| The algorithms and exploiture of a kind of small fingerprint identification system were mainly discussed in this thesis. The work of the thesis included preprocessing algorithm, minutiae extraction algorithm, minutiae matching algorithm of the image of fingerprints, and the method of setting up fingerprint minutiae database.At first, basic theories of fingerprint and fingerprint identification system were introduced synoptically. Then each part of fingerprint identification system was discussed in detail and some improved algorithms were put forward. In the part of image preprocessing, a segmentation algorithm for fingerprint image based on the information of gradient orientations was programmed. According to the uniformity and partial parallarity of ridges and furrows of fingerprint, a model of Gabor filter was established to enhance fingerprint image. Since two-dimensional Gabor filter could be separated into two one-dimensional filters, the complexity of the algorithm could be reduced efficiently. Based on the analysis of quick thinning algorithm and improved OPTA thinning algorithm, a synthesized thinning algorithm was presented. In the part of minutiae extraction, a kind of 8-neighbour extraction algorithm was proposed and distributing regulation of various pseudo minutiae in fingerprint image were analyzed. Special algorithms were designed with respect to various noises to delete pseudo minutiae. In the part of minutiae matching,the algorithm to acquire a pair of reference minutiae and transform parameters for fingerprint matching were programmed based on the relationship between two subsets that consist of three neighboring minutiae derived from the two minutiae sets. The technique of bounding box was combined with the algorithm to match fingerprint minutiae. Finally, a small automated fingerprint identification system was accomplished in the integrated software developing tool Visual C++ 6.0, and its capability was tested to be effective. |
PDF Full Text Request