No-reference Quality Metrics for Eye Fundus Imaging

摘要

This paper presents a comparative study on the use of no-reference quality metrics for eye fundus imaging. We center on auto-focusing and quality assessment as key applications for the correct operation of a fundus imaging system. Four state-of-the-art no-reference metrics were selected for the study. From these, a metric based of Renyi anisotropy yielded the best performance in both auto-focusing and quality assessment.