一种虹膜定位算法

An Iris Location Algorithm

针对虹膜二值边缘图像提取的困难 ,提出利用虹膜图像的灰度边缘图像以及虹膜的几何特征进行虹膜定位的快速算法 .首先根据虹膜图像的整体灰度分布信息特征 ,利用改进的 Sobel算子提取虹膜的灰度边缘图像 ,然后利用圆的对称性质以及投票策略提取虹膜的中心 ,最后利用 Hough变换的半径直方图投票提取虹膜内外径 .实验结果证明 ,该算法提高了虹膜定位的速度 ,并且具有较强的抗干扰能力 .

Owing to the difficulties in extracting iris edge image, a new iris location algorithm using geometric symmetry feature and edge detection from gray scale image is presented. Based on this idea, the improved Sobel detector is used to extract iris edge. Geometric features of circle and voting scheme are used to search the iris circle center, then the iris interior and exterior radii are detected by using radius histogram of Hough transform. Results show that the method can improve location speed and is not sensitive to noise effect.