一种基于指纹分类的指纹识别算法

依据指纹特征点在指纹图像中的特性给出了一种基于指纹分类的识别算法．该算法利用中心点和三角点信息对指纹进行分类并初匹配，然后用可靠性较高的分叉点所在脊线的相似度寻找出基准点对，根据基准点与周围特征点的关系筛选基准点对、确定变换参数，有效地解决基准点筛选受噪声影响问题，最后利用可变限界盒实现指纹匹配．实验结果表明，该算法能快速、准确地定位基准点，正确有效地实现指纹匹配，同时提高匹配速度及精确度．

A Fingerprint Verification Algorithm Based on Fingerprint Classification

According to the properties of fingerprint minutiae in fingerprints,this paper proposed a fingerprint verification algorithm based on fingerprint classification.The core and delta information is used to classify the fingerprint,and then the bifurcation ridge comparability is used to find reference minutiae according to the comparability of two ridges on which the highly reliable bifurcation lied.Meanwhile,in order to solve the problem that the reference minutiae filtering would be influenced by noise,taking the advantage of the relationship between the reference minutiae and the neighboring minutiae to filter the final reference minutiae and determine the transform parameters.And finally the algorithm implemented the fingerprint matching using the variable bounding box.The experiment results indicated that this algorithm had high speed and accurateness in locating the reference minutiae,more efficiency in performing fingerprint matching,as well as fastness and precision.