Abnormality detection in automated mass screening system of diabetic retinopathy

An approach of abnormality detection from color fundus images for automated mass screening system is proposed in this paper, which uses the object-based color difference image. Four color models, i.e. RGB, Luv, Lab and HVC are evaluated based on the hand labeled feature maps, and Luv and Lab are selected for computing color difference because of their good performance of object classification. The object-based color difference image of bright objects, e.g. exudates and drusen and dark objects, e.g. hemorrhages and blood vessel are obtained respectively according to the 2D histogram distribution on L-u plane, and then watershed transform is performed on the color difference image to extract object candidates. A pre-thresholding and a post-verification procedure are performed to deal with the over-segmentation problem of watershed transform.