渐进式虹膜图像质量评估模型

虹膜图像质量评估目前尚无统一评估标准，导致虹膜识别拒识率和误识率较高。针对该问题，提出一种虹膜图像质量评估模型。根据虹膜图像中各干扰因素的不同特点，在先验知识基础上采用区域化、加权的方法，渐近式地实施像素级质量评估，依据像素级评估结果实施图像级质量评估。实验结果表明，像素级虹膜图像质量评估中的虹膜图像干扰项识别率和模糊识别率较高，图像级虹膜图像质量评估与人工评估结果相一致。

Progressive Iris Image Quality Evaluation Model

Since there is no unified standard for iris image quality assessment at present, the recognition rejection rate and false acceptance rate cannot always be decreased efficiently. Aiming at this problem, this paper proposes a novel model for the assessment of iris image. According to the characteristics of different interfering factors in the iris image, regionalization and weighted methods are adopted on the basis of a prior knowledge in order to perform an incremental pixel-level evaluation of various interfering factors in iris images. The image-level overall evaluation based on pixel-level assessment results proceeds. Experimental results show that the pixel level iris image quality assessment has a better performance in blur recognition and interference term identification, image-level assessment of iris image quality assessment is consistent with the manual.