With the rapid development of computer and network technology, the intelligent identity authentication technology based on the recognition of biological features has received the widespread attention. In numerous biological recognition technology, fingerprint recognition technology is the earliest developed and the most widely used. In this paper,on the basis of the characteristics of the fingerprint minutiae in the fingerprint image, a recognition algorithm based on the topological theory and fingerprint classification is given. The algorithm is using the information of the core point and the delta point for fingerprint classification and match at the beginning, and then using the difference of the value of the orientation field between the bifurcation points with high reliability and the core point to find the reference points, at last, using the variable bounding box to match the fingerprint. In the experiments, the algorithm makes the matching speed improved by 40% , and decline the error rate and rejection rate about 0.5%. The results of the experiments show that, the algorithm can quickly, accurately locate the reference points, effectively solve the problem that the noise interferences the reference points screening, correctly and effectively match the fingerprint, and improve the matching speed and accuracy.%随着计算机和网络的迅速发展,基于生物特征识别的智能身份认证技术正受到越来越多的关注.由于指纹识别技术是生物识别领域技术中最成熟的一门应用技术,使得指纹识别成为目前应用最广泛、可信度最高的个人身份认证技术之一.文中依据指纹图像中细节特征点之间的关联性给出了一种指纹识别算法.该算法首先基于指纹分类学的思想,利用中心点和三角点的数量信息对待识指纹图像进行初匹配,然后基于拓扑学思想用可靠性较高的分叉点方向场及其与中心点的方向场差寻找出基准点对,最后利用可变限界盒实现指纹匹配.实验中,该算法使匹配速度提高了40％,误识率和拒识率略有下降,约0.5％.实验结果表明,该算法能快速、准确地定位基准点对,有效地解决提取基准点时的噪声影响,正确有效地实现指纹匹配,同时提高匹配速度及精确度.
展开▼
