The fingerprint identification has become one of the most popular, convenient and reliable personal authentication technologies due to its uniqueness and invariability. The fingerprint preprocessing, the minutiae extracting and the fingerprint matching are discussed in this paper according to the processing flow of the Automated Fingerprint Indentification System (AFIS).Based on the results of former researchers, we have done the research as follows:(1) In the aspect of fingerprint preprocessing, an improved preprocessing algorithm is proposed. In the process of the fingerprint enhancement, the Gabor filter formula is improved by pre-calculating the value of the two dimensional Gauss function. Then the computation is simplified from the multiplication of two dimensional Gauss function and Sin function to the multiplication of the pre-calculated value and Sin function, thus reduces the preprocessing time. When it comes to binarization, an improved binary method is proposed. We make full use of structural information, such as local fingerprint direction and fingerprint width, to extract the ridges and furrows from primitive fingerprint. Such method not only achieves binarization, but also enhances the fingerprint once again.(2) With regard to fingerprint matching, a kind of fingerprint matching algorithm based on bifurcation ridge comparability is proposed, which finds out possible reference minutiae according to the comparability of two ridges on which the highly reliable bifurcation lied. Meanwhile, in order to diminish the impact of noise on the selection of reference minutiae, we put forward a method which determines the final reference minutiae and transform parameters according to the relationship between the reference minutiae and the four neighboring minutiae. Lastly, matching of two fingerprints is accomplished by the bounding box of variable size. In a word, our method has increased the reliability of reference minutiae and guaranteed the correction of transform parameters.(3) Making use of the improved preprocessing algorithm and fingerprint matching algorithm based on bifurcation ridge comparability, the integrated fingerprint identification system concerned in our paper has been carried out. The results of experiment show that, our algorithms satisfy the design requirement, and make the... |