吉林大学学报(理学版) ›› 2018, Vol. 56 ›› Issue (5): 1156-1162.

• 计算机科学 • 上一篇    下一篇

基于形态学与灰度分布的序列虹膜质量评价算法

刘帅1,2, 刘元宁1,3, 朱晓冬1,3, 刘文滔1,3, 程垚松1,2
  

  1. 1. 吉林大学 符号计算与知识工程教育部重点实验室, 长春 130012;2. 吉林大学 软件学院, 长春 130012; 3. 吉林大学 计算机科学与技术学院, 长春 130012
  • 收稿日期:2017-11-08 出版日期:2018-09-26 发布日期:2018-11-22
  • 通讯作者: 朱晓冬 E-mail:zhuxd@jlu.edu.cn

Sequence Iris Quality Evaluation Algorithm Based onMorphology and Gray Distribution#br#

LIU Shuai1,2, LIU Yuanning1,3, ZHU Xiaodong1,3, LIU Wentao1,3, CHENG Yaosong1,2#br#   

  1. 1. Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China; 2. College of Software, Jilin University, Changchun 130012, China;3. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2017-11-08 Online:2018-09-26 Published:2018-11-22

摘要: 基于形态学和灰度直方图中的灰度分布, 提出一种虹膜质量评价算法. 通过清晰度, 基于二值形态学的活体检测, 灰度直方图灰度分布的区域有效判定以及超界偏移度和斜视离心度的斜眼程度判断, 解决了连续采集的序列虹膜质量评价中因判别指标严苛导致的有效虹膜幸存率过低的问题. 在排除不合格图像的同时, 尽可能地保留可以进行识别的虹膜, 以提高有效虹膜幸存率, 并且时间消耗相对较短.

关键词: 虹膜质量评价, 二值形态学, 区域有效判定, 超界偏移度, 斜视离心度

Abstract: We proposed an iris quality evaluation algorithm based on morphology and gray distribution of gray histogram. Through the sharpness, live detection based on binary morphology, region validity determination based on gray distribution in the gray histogram, and squint level based on ultraboundary deviation degree and strabismus eccentricity degree, it solved the problem of the low effective iris survival rate caused by  stringent discriminant index in continuous collection of sequence iris quality evaluation. While removing the unqualified images, the  proposed algorithm preserved iris which could be identified as far as possible, so as to improve the  effective iris survival rate, and time consumption was relatively short.

Key words: iris quality evaluation, binary morphology, region validity determination, ultraboundary deviation degree, strabismus eccentricity degree

中图分类号: 

  • TP391.41