Glaucoma Detection through image Processing

Abstract

Glaucoma is literally defined as a disease which commonly affects human eyes and it is also known as the second highest cause of blindness, worldwide. In order to avoid blindness, the condition needs to be treated at an early stage. Glaucoma is always associated with an increased intra-ocular pressure (IOP) in eye and it slowly leads to problems and totally collapses the vision of the patient. Ocular-hypertension is connected with the persons in whom IOP increases steadily and will not cause any damages to the optic nerves. Glaucoma shall classify as a. open-angle, b. close-angle, c. congenital, and d. normal tension based on its position of infection. The tension generated during glaucoma affects vision and damages optic nerve, ultimately ends up with vision loss. The automated analysis of retina images seems to one among the significant screening tool during the diagnosis process. This technique assists in detection of various disease and risks associated with eyes. Early diagnosis of this disease must be carried out in order to prevent the patients from permanent blindness. Screening of glaucoma is performed with its digital images of retina during the past few decades. There are few techniques are readily available to detect abnormality in the retina in reference to glaucoma. The significant processing techniques shall classed as image registration, fusion, segmentation, feature extraction, enhancement, morphology, pattern matching, image classification, analysis, and statistical measurements. The objective of this paper is to propose a technique which relates image processing and classification tools for reliable detection of glaucoma, compare and measure various parameters in fundus images obtained from the patients of glaucoma and those without the condition.