摘要
为了提高在图像质量较差时的虹膜定位性能,改善虹膜定位的成功率和速度,在对Daug-man算法改进的基础上,提出了一种基于形态学算子的虹膜定位算法(ILBM).该算法将睫毛干扰看作虹膜图像中出现的缝隙,通过应用形态学中的膨胀算子对缝隙进行填补进而消除干扰,虹膜边缘参数则采用Hough变换来求取.与Daugman算法相比,ILBM算法可以有效抑制虹膜定位过程中存在的睫毛干扰,提高虹膜定位的速度.对比实验表明,ILBM算法具有比Daugman算法更好的抑制干扰能力和鲁棒性,具有更高的定位成功率和更快的定位速度,定位时间仅为Daugman算法的17%,定位成功率提高了3.7%.
In order to improve the success ratio and velocity of iris location, an iris location algo- rithm based on morphological operators was developed. The iris location algorithm was realized through regarding the eyelashes as slits of iris image and applying the dilation operator of the morphology to fill up, and then the disturbance was removed. The Hough transform was used to get the parameter of iris boundary. So, the proposed iris location algorithm can reduce the errors caused by eyelashes and eyelids, and improve the speed of location process. Experiment validation has been carried out with iris image database. It shows that the improved algorithm can locate the iris more effective and quickly. Compared with the traditional algorithm of Daugman, and the location time is only 17% of Daugman＇s algorithm and the success ratio can be increased by 3.7%.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2007年第10期1175-1178,共4页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(60502021)
教育部高等学校博士学科点专项科研基金资助项目(20050698025)
关键词
边缘检测
虹膜定位
形态算子
edge detection
iris location
morphological operation
