A Novel Iris Image Quality Evaluation Based on Coarse-to-Fine Method

Ying Chen,Yuanning Liu,Xiaodong Zhu,Xiaoxu Zhang,Haiying Xu,Guang Huo,Ning Deng
DOI: https://doi.org/10.1007/978-3-319-02961-0_44
2013-01-01
Abstract:An effective image evaluation algorithm is vital for iris recognition system. This paper proposes a coarse-to-fine quality evaluation method to iris image. Five criterions are adopted in this paper, which are variance, gradient, edge strength, fuzzy entropy, information entropy, for iris image coarse evaluation. In fine quality evaluation phase, which mainly focus on evaluation of spatial location and effective regional. The experimental results show that our proposed method are able to meet the need of practical iris recognition system.
What problem does this paper attempt to address?