【作者】 张仁彦； 赵国良； 苏丽；

【Author】 ZhangRenyan, ZhaoGuoliang,SuLi(Automation College Harbin Engineering University, Harbin 150001

【机构】 哈尔滨工程大学自动化学院；

【摘要】 作为虹膜识别的关键步骤,虹膜定位由内、外边缘定位两部分组成。在虹膜内边缘定位中,该文分析并得出虹膜图像行、列灰度曲线几何特征与瞳孔区域边界位置之间的对应关系,即:虹膜图像行、列灰度曲线上与瞳孔区域对应的部分有类似于“深井”的几何特征。在这种对应关系的基础上,该文提出通过寻找“深井”最大宽度的方法定位瞳孔。在虹膜外边缘定位中,该文采用圆形边缘检测器进行虹膜外边缘定位。实验结果表明该方法能够准确地定位虹膜边缘。与其它几种虹膜定位算法相比,该文提出的方法运算量小、计算简单、速度快、受眼睫毛和眼睑的影响小、稳定性高。更多还原

【Abstract】 As a key stage in the iris recognition, the iris location includes the location of the iris inner boundary and the iris outer boundary. In the iris inner boundary location, it was analyzed and gotten that the relations between the geometrical characters of the row or column curves of the gray levels and the location of the pupil boundary in an iris image, viz. the sect corresponding to the pupil in the row or column curve of the gray levels had the geometrical character like a deep well. The paper proposed a method for locating the pupil by finding the maximal width of the deep well based on the relations mentioned above. In the iris outer boundary location, the iris outer boundary can be located by circle detector. The experiment results have shown that the proposed method can locate iris perfectly. Comparing with other iris location methods, the proposed method is with less calculation burden, simpler for calculation, faster, not easily influenced by eyelash or eyelid and more robust.更多还原