Abstract: A novel minutiae-based fingerprint matching method is proposed. First a new structure adjacent feature union（AFU） is defined， which is invariant to rotation and translation. The AFU （adjacent feature union） is used for fingerprint alignment. Moreover， the ridge frequency and block orientation are utilized to construct a distortion-tolerant model. Using this model， the position and direction of the minutiae can be readjusted and thus the performance of the matching can be enhanced. The proposed method can minimize the effect of the elastic distortion in fingerprint. Experiment results demonstrate the high effectiveness of the matching method.

Key words: fingerprint matching, minutiae, elastic distortion, adjacent feature union （AFU）

摘要： 提出了一种基于细节点的指纹匹配方法。定义了一种新的结构邻接特征联合体（AFU），并用这个与旋转和平移无关的局部特征与指纹细节点进行比对；利用纹路的频率和块方向信息对细节点的位置和方向进行重新调整以增加匹配的可靠性。实验结果表明该方法可以很好地处理指纹中出现的形变问题，具有较好的匹配效果。

关键词: 指纹匹配, 细节点, 弹性形变, 邻接特征联合体