Classification of Hemorrhages for Diabetic Retinopathy Screening

Advances in the eye care telemedicine system aid the diabetic patients in remote areas to stop the unwanted visit to the ophthalmologist, reduces the overall time and money. Diabetic retinopathy, which causes sight loss, has the most common symptoms like microaneurysms, hemorrhages, cotton-wool spots, exudates and drusen. In this work, an efficient approach for the automatic detection of hemorrhages in color retinal images is proposed and validated. The color retinal images captured from the diabetic patients are enhanced by an effective Otsu method, Adaptive Histogram Equalization, K-means algorithm. A bag of features based on intensity, color and texture are extracted. Finally, the features are classified with the help of SVM with FUZZY classifier. The classifier performance is validated on two publicly available datasets (DIARETDB1, MESSIDOR).