计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (1): 201-206.DOI: 10.3778/j.issn.1002-8331.1511-0108

• 图形图像处理 • 上一篇    下一篇

一种结合Snake与GAC模型的虹膜定位方法

凌  滨,索健文,许景涛   

  1. 东北林业大学 机电工程学院,哈尔滨 150040
  • 出版日期:2017-01-01 发布日期:2017-01-10

An iris location method based on Snake and GAC model

LING Bin,SUO Jianwen,XU Jingtao   

  1. College of Machinery Electricity, Northeast Forestry University,Harbin 150040,China
  • Online:2017-01-01 Published:2017-01-10

摘要: 为了提高虹膜定位的精度和可靠性,采用改进的Snake模型和改进的GAC模型分别定位虹膜的内外边缘。对于虹膜内边界,在传统的Snake模型的基础上,引入GVF力场代替原有的传统外力场,完成Snake模型对虹膜内边缘的定位。对于虹膜外边界,在传统的GAC模型的基础上,引入非负常数项作为演化收缩力,使其方向指向对象内部,从而实现GAC模型对虹膜外边缘的定位。采用CASIA虹膜库进行了大量实验,结果表明该方法与经典的虹膜定位算法相比,定位速度更快,抗噪性更强,定位更准确。

关键词: 虹膜定位, 活动轮廓, Snake模型, GAC模型

Abstract: In order to improve the accuracy and reliability of iris location,this paper uses the improved Snake model and the improved GAC model to locate the inner and outer edge of the iris. On the basis of the traditional Snake model,the GVF force field is used to replace the traditional external force field to locate the inner edge of the iris. On the basis of the traditional GAC model,the non negative constant term is introduced as an evolutionary contraction force,which makes the direction of?the object inside,and realizes the location of iris outer edge by GAC model. A large number of experiments are carried out using CASIA iris database. The results show that the method is faster,more accurate positioning,stronger in restraining noise than the classical iris localization algorithm.

Key words: iris location, active contour, Snake model, GAC model