Abstract: Against low-quality images，a method to locate accuratly fingerprint image center point is proposed by complex filter，and in order to overcome the shortcomings of finding fingerprint center point by traditional automatic fingerprint identification.At the same time，an algorithm of feature extraction and matching based on the contour is proposed，the algorithm using the overall structure feature of fingerprint and improving the correct rate of matching.The experimental results show that this algorithm has a very high recognition rate and robust.

摘要： 针对传统指纹细节点提取和匹配方法在处理低质量图像上的不足，首先提出了一种利用复滤波器准确定位指纹图像中心点的方法，克服了传统自动指纹识别中寻找指纹中心点不够精确的缺点。同时提出了一套基于轮廓的指纹特征提取和匹配算法，此算法利用指纹的总体结构特征，提高了匹配的正确率。实验结果表明，该算法具有相当高的识别率和较强的鲁棒性。