基于SIFT和SDM的虹膜定位方法

Iris Localization Method Based on SIFT and SDM

为了提高虹膜定位的速度和稳定性,提出一种基于SDM的快速、稳定的虹膜定位算法.该方法首先采用径向对称变换粗定位瞳孔,然后采用微积分算子精定位瞳孔;选取SIFT特征描述虹膜外边缘及眼睑的边界特征,采用SDM算法求解定位结果,最后采用最小二乘法确定虹膜外圆及上、下眼睑边界参数.实验结果表明该算法大大提高了虹膜定位的效率和稳定性.

To improve the speed and stability of iris localization,an SDM（supervised descend method）-based iris localization algorithm was proposed. Firstly, radial symmetry transformation was adopted to localize pupil roughly, then, an integro-differential operator was used to segment pupil accurately. Secondly, SIFT （ scale invariant feature transform） features were selected to describe the characteristic information of the outer boundary of the iris and eyelids. Thirdly, the SDM algorithm was employed to determine the key points on the outer boundary and eyelids. Finally, the least square algorithm was used to determine the parameters of the outer boundary, the upper and lower eyelids. Experimental results show that the proposed algorithm can greatly improve the efficiency and stability of iris localization.