Discriminant color texture descriptors for diabetic retinopathy recognition

Abstract

Diabetic retinopathy (DR) is a common eye disease that could lead to irreversible vision loss but hard to be noticed by carriers in early stages. Instead of isolating DR signs for DR recognition, this paper examines discriminant texture features obtained by color multi-scale uniform local binary pattern (LBPs) descriptors on five common color spaces and two proposed hybrid color spaces. The extracted features are evaluated by the enhanced Fisher linear discriminant, EFM. Experiments are done on a large dataset of 35,126 training images and 53,576 testing images that have been taken by different devices with high variance in dimensions, quality and luminance. The best performance is above 71.45% by HSI-LBPs, a*SI-LBPs, and bSI-LBPs descriptors.