基于虹膜识别技术的身份验证模型

Authentication model based on iris recognition technology

文章从虹膜识别的实际应用着手,重点分析特征编码的存储和海明距阈值的选定,以提高匹配精度,缩短匹配时间。身份验证模型中,所有实验虹膜图像均来源于CASIA-Iris-Syn图像库,已注册用户的虹膜特征信息将保存在xls文件中;同时计算已注册的虹膜图像与其所属个体中所有图像的海明距离平均值,并以这些平均值最大值再一次与0.3一起求得的平均值作为阈值来匹配新虹膜图像。实验证明,这种方法能有效地提高匹配速度,降低错误接受率。

Starting with the practical application of the iris recognition, this paper focused on analysis of feature code storage and Hamming distance threshold selected to improve the matching accuracy, and shorten the matching time. In this authentication model, all of the testing iris images were from CASIA-Iris-Syn; the registered user＇s iris feature information was saved in the XLS file; the average of Hamming distance between the registered image and all the other images which belong to the same person, was calculated firstly, and then the max of them would be used to take average with 0.3 together, this new average would be the threshold to match a new iris image. The experimental results show that this method can effectively improve the matching speed and reduce the false acceptance rate.