RETINAL BLOOD VESSEL SEGMENTATION USING ROI DETECTION AND PCA CLASSIFICATION

Retinal blood vessel segmentation is an advanced technique being used for the detection of several eye diseases. This project proposes a strategy for the identification of diseases like diabetic retinopathy, glaucoma, mascular degeneration etc., through the discovery of exudates. Exudates is the lipoprotein that gets leaked out of the damaged blood vessels of a human eye. These exudates, damaged vessels are extremely difficult to be identified by visual inspection. An efficient image analysis program can be used to detect their presence effectively. In this project we have proposed one such method where the disease can be identified for its presence using the fundus image of an eye. The image is then pre-processed and it helps ophthalmologists in the detection of diabetic retinopathy disease using region of interest-based segmentation and a principal component analysis algorithm being implemented for the classification of disease.