Automated fundus image field detection and quality assessment

摘要

A method, system, and computer readable medium which automatically determine the side, field and a level of image quality of fundus images of the retina of a human eye is disclosed. The disclosure combines image processing, computer vision and pattern recognition techniques in a unique way to provide a robust process to identify and grade the quality of fundus images with application to improve efficiency and reduce errors in clinical or diagnostic retinal imaging workflows.