中文摘要：

为了提高虹膜识别系统的识别性能,针对虹膜识别中的特征提取与模式分类问题,提出了一种基于核Fisher鉴别分析（kernel fisher discriminant analysis,KFDA）与支持向量机（support vector machine,SVM）的虹膜识别方法。从采集到的人眼图像中定位虹膜,并对其进行归一化处理;使用核Fisher鉴别分析提取虹膜纹理特征,并通过选择合适的特征个数提高识别的准确率;在得到虹膜特征编码后,用支持向量机进行分类判决。对CASIA虹膜库的测试结果表明,该方法的处理速度是Daugman虹膜识别方法的4.4倍;该方法与Boles虹膜识别方法相比,降低了错误接受率和错误拒绝率。实验结果表明：该方法能更好地提高虹膜的识别率和降低虹膜识别时间。

英文摘要：

A new method was proposed to improve the performance of iris recognition,which was named iris recognition method using kernel fisher discriminant analysis（KFDA） and support vector machine（SVM）.Firstly,normalization was used to process the iris position which was located in the eye images.And then KFDA was used to extract statistical independent feature and a good result would be received by selecting right feature numbers.The SVM was used to classify the iris feature.Experiment results on the CASIA iris database showed that the new method could handle iris data sets more quickly compared to the Daugman′s method and have decreased the false acceptance rates and the false rejection rates than Boles′s method.The experimental result showed that the proposed method was effective and feasible.It also had high recognition accuracy and speed.