指纹图像预处理和特征提取  

Preprocessing and Minutiae Extraction of Fingerprint Images

作　　者：张正言[1] 戴晓强[1] 张冰[1] 

机构地区：[1]江苏科技大学电子信息学院,江苏镇江212003

出　　处：《微电子学与计算机》2008年第7期134-136,141,共4页Microelectronics & Computer

摘　　要：在对现有的指纹图像增强算法研究和分析的基础上,结合指纹图像的方向特性和灰度特性,提出了一套较完整的指纹图像预处理和特征提取算法.先利用Kirsch算子得到指纹图像的方向场,并对指纹图像进行方向中值滤波,然后通过图像的边缘强度设计局部自适应阈值对指纹图像进行二值化,最后在指纹的细化图像上提取出指纹的细节特征点,并对其进行去伪处理.实验结果表明,该方法处理效果好,运算速度快.Based on investigation and analysis of the arithmetic in existence, a set of algorithm for preprocessing and minutiae extraction of fingerprint image is proposed, which integrated with the direction and gray level statistics of the fingerprint image. Firstly gain the directional image of the fingerprint based on the Kitsch operator, and filter the fingerprint images using the algorithm named directional raedial filtering, secondly obtain the binarization images after segmenting the fingerprint images by the local adaptive thresholds designed by the edge intensity of images, and lastly extract minutiae of a fingerprint from its thinned image and delete spurious minutiae. The experiment results show that this algorithm has the advantages of good processing effect and work fast.

关 键 词：KIRSCH算子 滤波 边缘强度 二值化 特征提取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]