吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

基于极坐标特征的改进模糊C-均值虹膜定位算法

王洋1,2, 庞彦尼3, 左平4   

  1. 1. 吉林大学 计算机科学与技术学院, 长春 130012; 2. 吉林大学 公共计算机教学与研究中心, 长春 130012;3. 吉林大学 数学学院, 长春 130012; 4. 空军航空大学 基础部, 长春 130022
  • 收稿日期:2013-04-07 出版日期:2014-05-26 发布日期:2014-08-27
  • 通讯作者: 庞彦尼 E-mail:pangyn@jlu.edu.cn

Improved FCM Algorithm Based on Polar Coordinates Transformation for Iris Location

WANG Yang1,2, PANG Yanni3, ZUO Ping4   

  1. 1. College of Computer Science and Technology, Jilin University, Changchun 130012, China;2. Center for Computer Fundamental Education, Jilin University, Changchun 130012, China;3. College of Mathematics, Jilin University, Changchun 130012, China;4. Department of Foundation, Aviation University of Air Force, Changchun 130022, China
  • Received:2013-04-07 Online:2014-05-26 Published:2014-08-27
  • Contact: PANG Yanni E-mail:pangyn@jlu.edu.cn

摘要:

针对非理想情况下虹膜图像定位失败的问题, 提出一种新的虹膜定位算法. 该算法先使用基于极坐标特征的改进模糊C-均值算法对虹膜外圆半径进行粗定位, 再采用圆周差分法对外圆参数进行准确计算. 该算法通过使用极坐标作为聚类特征及放宽模糊聚类的聚类条件, 提高了虹膜定位算法的鲁棒性. 实验结果表明, 该方法有效提高了非理想情况下虹膜图像的定位精度.

关键词: 虹膜定位, 虹膜鉴别, 极坐标变换, 改进的模糊C-均值

Abstract:

A new twostep location approach combing coarse location with fine location was presented for iris location under nonideal situation. Improved fuzzy C-mean clustering based on polar coordinates transformation was proposed for the coarse iris location and a kind of dynamic circular edge template was adopted by the fine iris location. Having used polar coordinates as clustering attribute and relaxed fuzzy clustering restriction, the proposed algorithm improves the robustness of iris location. The experimental results prove the validity of this approach convincingly.

Key words:  iris location, iris identification, polar coordinates transformation, improved fuzzy C-mean

中图分类号: 

  • TP391.1