About

INTRODUCTION

Eye is very important in our daily life that they gives which allows us to see dimensions of objects, shapes and colors by processing the light they reflect or emit. Retina is a part of the eye.Since most of the problems leading to vision loss occur are sources from retina, it is important for an ophthalmologist to be able to do an accurate diagnosis in so that any abnormality in retina can be detected earlier. The retina is the only location where blood vessels can be directly captured. Eye diseases such as glaucoma and diabetic retinopathy are one of disease related to retina that serves as the main cause of blindness. Even though there are many different techniques have been proposed in order to detect the blood vessels and lesion in retinal image, there is still some issue on how accurate and fast the method works. Blurred retinal image may be mistakenly as lesions graded as normal. Therefore a sufficient clarity of blood vessel extraction is needed for the retinal image. Furthermore, studies have shown that people who suffer from diabetes benefit from regularly attending a screening session. Therefore a better method for detecting a blood vessel must be search since early detection of disease may prevent further damage to patient’s eye thus reducing the probability of blindness. This project is proposed to be able to identify the blood vessel and lesion in retinal image so that any abnormality behavior can be detected earlier. The sample retinal image is obtained from DRIVE and STARE database.

OBJECTIVES

 PROBLEM STATEMENT

DATABASE