摘要
文章针对指纹识别系统存在的问题提出了一种利用指纹的结构特征和指纹线特征相结合确定参照点（极坐标原点）的方法，克服了传统自动指纹鉴别中寻找指纹中心点不够精确和耗时过多的缺点，同时利用极坐标变换将指纹特征矢量化，很好地解决了指纹图像本身非线性时变等因素对指纹识别的影响，算法具有相当高的识别率和较强的鲁棒性。
A novel technique for automed fingerprint authentication is presented which utilizes the structural features and polar co-ordinate transformation to promote the rate of.fingerprint recognition.By combining the use of those two techniques,authors have developed a unique multilevel verification/identification system which has a high rate of fingerprint recognition and creates a more strong robust overall set of features.Not only is a special algorithm of image preprocessing given to enhance fingerprint ridge structure, but an adaptive elashc matching algorithm is also given to compensate the minutia localization errors and nonlinear deformations.That low false reject and zero false accept error rates have been observed based on initial testing of the prototype verification system.
出处
《计算机工程与应用》
CSCD
北大核心
2000年第6期33-35,59,共4页
Computer Engineering and Applications
关键词
指纹识别
极坐标变换
指纹图像
图像处理
fingerprint recognition, structural feature, polar co-ordinate system,feature vector, linear matching
