摘要
采用指纹增强算法与复滤波器中心点定位结合的方法,有效地解决低质量指纹图像中心点定位问题,实现了平移不变.同时,采用基于结构的指纹特征提取算法,用八方向Gabor滤波提取指纹的全局特征和局部脊线特征,实现了旋转不变.在FVC2004指纹库上的仿真实验表明:该算法取得较好的识别效果,优于基于单Finger Code特征和基于点模式的指纹识别算法.
This paper solves the center point localization on poor quality fingerprints effectively, by using the fingerprint image enhancement algorithm coupled with the localization of core point by the complex filters, and realizes the translation invariant. In addition, the eight directions Gabor filter can capture the global and local ridge structures by using the method on structure-based fingerprint feature extraction, and realize the rotation invariant. The simulation experimental results based on the FVC2004 fingerprint database show that the performance of the paper is better than that of the single finger code-based and the minutiae-based method.
出处
《华侨大学学报（自然科学版）》
CAS
北大核心
2013年第2期151-155,共5页
Journal of Huaqiao University(Natural Science)
基金
福建省自然科学基金计划资助项目(2010J01340
2012J01275)
关键词
指纹识别
图像增强
中心点
GABOR滤波
特征匹配
fingerprint recognition
image enhancement
core point
Gabor filter
feature matching
