Efficient optic cup localization based on superpixel classification for glaucoma diagnosis in digital fundus images

Optic cup is the primary image indicator clinically used for identifying glaucoma. To automatically localize the optic cup in fundus images, an effective and efficient superpixel classification based approach is proposed in this work, which maintains both advantages of existing pixel and window based approaches. This method provides three major contributions. First, it proposes processing of the fundus images at the superpixel level, which leads to more descriptive and effective features than those employed by pixel based approaches, without additional computational cost. Second, a feature normalization method is proposed to reduce the influence of illumination variations across the training and testing images, which greatly elevates superpixel classification accuracy. Third, a refinement scheme that utilizes both retinal structural priors and local context information is adopted to further improve the accuracy. Tested on the ORIGA-light clinical dataset, which comprises of 325 images from a population-based study, the proposed method achieves an accuracy that is comparable to or higher than the state-of-the-art techniques, with a speedup factor of tens or hundreds.