非接触式虹膜图像获取及特征提取方法的研究

报道了一种由使用者自行瞄准来实现虹膜图像非接触式获取的新方法，对获得的图像数据进行了虹膜纹理特征提取试验，验证了方法的可行性和有效性。通过设计几何光照明来辅助使用者定位瞄准，使眼睛处于前后合适的位置；并在摄像机镜头前面放置一片半透半反镜，帮助使用者观察自己眼睛的像来判断眼睛是否偏出摄像机的视场范围。通过该方法可以得到虹膜特征识别的图像数据。采用二维伽博小波对虹膜纹理进行特征提取，并计算两个虹膜特征码值中数值相等的特征位个数，对采集到的虹膜图像进行了特征匹配实验，得到的实验数据证实了该系统完全满足虹膜特征识别的要求。

Research on Non-Contact Method of Capturing Iris Image and Extracting Feature

A new method for iris recognition is presented, which can help users to get correct position and accomplish non-contact capture by self-aiming. The experiments on extracting feature from iris texture are carried out, and the feasibility and efficiency of this method are verified. The special illumination is designed for the purpose of assisting users to focus their eyes on the proper position. A piece of pellicle mirror is placed in front of the camera lens to help users to observe the eyes' image and ensure that the eyes are in the camera visual field. Thus image data that meet the requirement of iris recognition can be obtained. The two-dimensional Gabor wavelet theory is adopted for extracting feature from iris texture. Feature matching experiments are made by counting the amount of the equal flag bits of two iris images.