基于方向图的动态阀值指纹信息提取识别研究

摘要

Research accurate recognition of fingerprint. Fingerprint structure is complex and vulnerable to interference. This paper put forward a fingerprint identification algorithm based on dynamic threshold value in orientation images. The topological structures of noises in fingerprint images were analyzed, then filtering of image was carried out. New direction figure dynamic threshold filtering was used and binary fingerprint image was obtained. The experiment results show that the improved algorithm can improve the accuracy of identification.%研究了指纹身份信息识别优化问题.由于在进行指纹识别时,指纹图像容易被噪声干扰,针对传统的指纹识别算法由于噪声的干扰,影响了指纹特征提取准确度.为了有效的解决上述问题,提出了一种新的方向图动态阈值指纹身份特征提取识别算法.首先分析指纹图像中存在的各类噪声的拓扑结构,然后对指纹图像预滤波处理,采用新的方向图动态阈值图像滤波和指纹图像二值化处理,最后得到二值化细化处理后的指纹图像.仿真结果表明,改进算法是一种有效的滤除指纹细节伪特征点的方法,可以有效的提高指纹信息提取识别的准确度.