计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (26): 197-199.

• 工程与应用 • 上一篇    下一篇

一种虹膜图像质量评价的方法

卢光明1,2,戚佳音2,廖庆敏1   

  1. 1.清华大学 深圳研究生院,广东 深圳 518055
    2.哈尔滨工业大学 计算机科学与技术学院,哈尔滨 150001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-09-11 发布日期:2007-09-11
  • 通讯作者: 卢光明

Iris image quality assessment method

LU Guang-ming1,2,QI Jia-yin2,LIAO Qing-min1   

  1. 1.Graduate School at Shenzhen,Tsinghua University,Shenzhen,Guangdong 518055,China
    2.School of Computer Science and Technology,Harbin Institute of Technology,Harbin 150001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-11 Published:2007-09-11
  • Contact: 卢光明

摘要: 虹膜图像质量是影响虹膜识别系统精度的关键。从序列图像中挑选出符合要求的虹膜图像能有效地降低识别系统的拒识率和误识率。这里按照产生虹膜图像奇异的不同情形:失焦、运动模糊、眼睑遮挡及睫毛遮挡,充分利用虹膜自身的纹理分布特点,在算法中引入了评价区域局部化、小波包分解和能量加权等思想,分别设计相应算法加以评判。实验结果表明该评价方案快速有效,其评价结果和人眼主观评价相吻合。

关键词: 虹膜识别, 质量评价, 小波包

Abstract: The quality of iris image is the key point to affect the accuracy of iris recognition system.Selecting good iris images from a sequence frames can efficiently depress the false rejection rate and false acceptance rate of the recognition system.Four main causes from various factor-oriented noises often degrade iris image quality:defocus,motion blur,eyelid occlusion,and eyelash occlusion.Considering the texture distribution characteristics on iris images,an image quality assessment method is proposed corresponding to different situations based on the ideas of evaluation area localizing,wavelet packet transform,and weighted energy.The experiment results show that the presented method is efficient and real-time in assessing the quality of iris images,and the evaluation results are coincident to the judgment of human eyes.

Key words: iris recognition, image quality assessment, wavelet packet transform