【摘要】 为改善虹膜定位的效果,提出一种改进的基于Hough变换的虹膜定位算法。首先,用一个全1的矩形窗对瞳孔中心进行估计;然后,以该估计中心为极点对虹膜二值图像进行极坐标变换,通过水平边缘点选择规则剔除非水平边缘点,再将图像变换到直角坐标;最后,利用Hough变换以及虹膜内、外边界之间存在的耦合关系来求取虹膜边界参数,并取最大和次大参数的均值作为最终的边界参数。实验结果表明,该算法平均定位时间为0.152 s,准确率高达98.4%。更多还原

【Abstract】 The purpose of this paper is to research an improved algorithm of iris localization based on Hough transform, for improving the effect of iris localization. Firstly, the pupil center is estimated by using a total of 1 rectangular window. Then, the binary edge image is transformed into rectangular region which is indicated with polar coordinates, and the rules of horizontal edge selection are used to remove the non- horizontal edge, and then the edge- selected image is transformed into cartesian coordinates. Finally, the Hough transform and the coupling relationship between the internal boundary and the external boundary are employed to determine the parameters of boundaries of the edge- selected iris image, and the mean of the biggest and the second biggest bound-ary parameters are chosen as the estimate of the iris boundary parameters. Experiments indicate that the proposed algorithm costs an average time of 0. 152 s, and the correct rate is 98. 4 %.更多还原