摘要
为了提高虹膜识别系统的性能,提出了一种新的虹膜识别算法。首先采用先粗后细的方式定位虹膜的内外边界,然后将虹膜图像归一化展开并去除噪声,最后用选定的滤波器虚部对展开后的虹膜图像进行滤波并对特征编码,匹配算法中采用Hamming距离来定义特征间的距离。实验表明:对于CASIA库中的图像及自制采集设备采集的图像,本文算法都能达到较好的识别效果,而且相对于传统算法,本文算法大幅度降低了处理耗时和虹膜特征编码长度。
In order to improve the performance of iris recognition,a novel iris recognition method is presented.Firstly the algorithm locates the inner-circle and outer-circle of iris,then it expands and denoises the iris.Secondly it filters the iris image and encodes iris feature.In the matching method,the distance of two features is defined by Hamming distance.Experimental results show that this algorithm is effective for iris images and it shortens the processing time and code-size of iris feature by a large margin...
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2009年第11期1507-1510,共4页
Journal of Optoelectronics·Laser
基金
国家自然科学基金资助项目(30570473)
