基于人眼灰度分布特征的虹膜定位算法

虹膜定位是虹膜识别过程中的重要环节，定位速度和精度决定了整个虹膜识别系统的性能。提出了一种基于人眼灰度分布特征的虹膜定位算法，该算法利用形态学运算实现瞳孔圆心粗定位，采用划分区域求灰度均值隔项差值最大值的方法实现外圆半径的粗定位，并通过分层聚类的方法实现内外边界的精确定位。实验结果表明．与经典的虹膜定位算法如Wildes算法、Daugman算法相比，该算法定位结果更加准确，定位速度大幅度提高。

An Iris Location Method Based on Gray Distribution Features of Eyes

Iris location is a kernel procession in an iris recognition system. The performance of the iris recognition system is determined by the speed and accuracy of the iris location decide. An iris location method based on gray distribution features of eyes is presented. This method uses morphologic algorithm to calculate the pseudo-centre of pupil, then divides the area around pupil to several annular region to calculate the gray mean to find the maximal difference alternately to decide the pseudoradius of the outer boundary, at last uses layered clustering algorithm to accurately locate the two edges. The experimental results demonstrate that the location results of proposed algorithm are more accurate and rapid than any other classical al gorithms such as Wildes＇ algorithm and Daugman＇s algorithm.