用于虹膜分类和旋转校正的角点特征预处理方法

提出了基于角点特征的虹膜预处理方法。该方法利用虹膜区域中角点及邻域灰度信息,通过仿射对应并进行相似度计算判别模板和待识图像间的类属关系。解决了样本集划分和精细识别中旋转角度去除(保持旋转不变性)的问题。实验结果表明:本文方法具有很好的识别分类性能和较快的识别速度,可作为完成降低搜索规模及去旋转归一化的预处理操作。

Preprocessing method for iris classification and rotation correction based on corner characteristics

An iris preprocessing method was proposed based on corner characteristics.Using the information of corner and gray-scale in the corner neighborhood region,the type of relationship between the template image and identifying image is determined through the corresponding affine and similarity calculation.The method solves the problem of removing rotation(maintaining rotation invariance) in the division of the iris sample set and fine recognition.Experimental results indicate that the proposed method has good classification performance and rapid recognition speed.It can be taken as the initial preprocessing operation for reducing the search size and removing the rotation in the normalized phase.