Font Size: a A A

Image Quality Assessment For Iris Recognition Technology Research

Posted on:2011-09-16Degree:MasterType:Thesis
Country:ChinaCandidate:F X ShiFull Text:PDF
GTID:2178360305469914Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
As an emerging biometric technology, the iris recognition has received wide attention from scientists and researchers worldwide. Compared with other biometric technology, the iris recognition has many advantages, such as uniqueness, stability, security, non-invasive. Therefore, iris recognition technology is considered one of the mainstream biology recognition technologies in future and a hot realm of research.The recognition rate of iris recognition system is severely affected by poor-quality images. In this paper, a progressive iris image quality evaluation model is proposed for the problem above. The main contents of this paper include:(1) Through the analysis of the Daugman iris recognition system and Wildes iris recognition system, the requirement model of iris image quality evaluation for iris recognition system is clear;(2) Through the comparison and the analysis of the existing iris image quality evaluation technology, we propose a progressive iris image quality evaluation model based on the foundation of the predecessors' outstanding achievements. The proposed model carries out the progressive quality evaluation from both the pixel level and the image level. The pixel level evaluation includes such aspects as illumination evaluation, reflection evaluation, localization rationality evaluation, fuzzy evaluation and mask evaluation;(3) To verify the iris picture quality evaluation model in this paper, we combined the proposed model and the Masek iris recognition system. The iris images of CASIA, UBIRIS, WVU iris library were used for the experiments. And a comparison of the results between the proposed model and some predecessors'models was made.Experiments show that the proposed evaluation model is superior to or as good as existing methods in single-factor evaluations such as fuzzy evaluation and block evaluation, and the model proposed in this paper assesses the factors that are more universal. The model better solves the issues of iris image quality evaluation. It has a certain theoretical and practical value.
Keywords/Search Tags:Image Quality Assessment, Quality Assessment of Iris, Iris Recognition, Wavelet Packet Decomposition
PDF Full Text Request
Related items