基于局部细节邻接图的指纹匹配算法研究

Fingerprint matching algorithm based on local minutiae adjacency graph

指纹匹配是指纹识别的关键环节,它根据指纹的特征描述方式来判定两幅指纹图象是否来自同一个人的同一手指,它的好坏直接反映了指纹自动识别系统的最终效果.本文基于以指纹的细节特征为顶点构造的局部细节邻接图,提出了一种分阶段的指纹匹配算法.相对于现有匹配算法,它的由局部到全局、分阶段的匹配标准更加可靠、优越.经实验证明该算法具有较快的速度和较高的准确率,可应用于实际的指纹自动识别系统中.

Fingerprint matching is the critical step in fingerprint identification processing. Based on the predefined mode of fingerprint features, it identifies whether two fingerprint images came from the same finger. It intances in a fingerprint image. Proposes a new integrated-match method based on local minutiae adjacency graphs （LMAG） with the minutiae features as the nodes. Comparing with the existed fingerprint matching algorithms, the new algorithm implements the match processing from local to global, which is more reliable and robust. The experimental results show that the algorithm is fast and has high accuracy. It can be applied to the practical fingerprint automatic identification system.