Computer Engineering and Applications ›› 2018, Vol. 54 ›› Issue (18): 212-217.DOI: 10.3778/j.issn.1002-8331.1706-0022

Previous Articles     Next Articles

Iris location algorithm based on partitioning search

LIU Shuai1,2, LIU Yuanning1,3, ZHU Xiaodong1,3, FENG Jiakai1,3, LU Shixu1,2   

  1. 1.Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China
    2.School of Software, Jilin University, Changchun 130012, China
    3.School of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Online:2018-09-15 Published:2018-10-16

基于分块搜索的虹膜定位算法

刘  帅1,2,刘元宁1,3,朱晓冬1,3,冯家凯1,3,卢时旭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

Abstract: Aiming at the problem that the traditional iris localization algorithms are not stable and low robustness, iris location based on partitioning search is proposed. Firstly, according to the difference of the gray scale of iris image, iris image is transformed into binary image, and inner circle is rough positioned by using the hough circle detection method based on the edge detection, then through using partitioning search of binary image, the inner circle is accurately located. Using convolution operation to rough locate the iris outer circle, and then partitioning search is used for original image, the gray scale change in the gray scale histogram is observed to precise position the iris outer circle. The obtained iris and the iris got from traditional iris localization algorithms are processed by the same iris recognition algorithm. The result shows that this algorithm has better effect on image recognition and has good robustness.

Key words: partitioning search, iris location, hough circle detection, convolution operation, gray scale histogram

摘要: 针对传统虹膜定位算法识别效果不稳定,鲁棒性低的问题,提出基于分块搜索的虹膜定位算法。首先利用虹膜图像灰度变化差异将虹膜图像转换为二值图像,用基于边缘检测的hough圆检测法粗略定位虹膜内圆,再利用分块搜索二值图像对内圆进行精确定位。之后利用卷积运算粗略定位外圆,再对原图像进行分块搜索,观察截图的灰度直方图中的灰度变化精确定位外圆。将得到的虹膜与传统定位算法得到的虹膜用相同的虹膜识别算法处理,结果表明,该算法定位出的图像识别上效果更明显,并且具有很好的鲁棒性。

关键词: 分块搜索, 虹膜定位, hough圆检测, 卷积运算, 灰度直方图