A SURVEY ON MACHINE LEARNING BASED TECHNIQUES FOR DETECTION OF GLAUCOMA

Glaucoma is often termed as the silent snatcher of eyesight. It is one of the leading causes for blindness which is caused due to abnormally high pressure created in the eye, causing damage to the optic nerves. With the advancements in image processing and machine learning techniques, automated detection of glaucoma has been an active area of research. The retinal fundus images is analysed for detection of glaucoma which is critical to prevent loss of vision. An early detection of glaucoma can help to arrest the condition and avoid further degradation to the optic nerves and vision quality. This paper analyses the various techniques pertaining to retinal image processing, noise removal, segmentation, feature computation and classification. Different machine learning based classifiers are evaluated in terms of the performance metrics such as accuracy, sensitivity, precision and recall. A comparative analysis of the noteworthy contribution in the field is also presented which can lay the foundation for the development of novel algorithms to address the challenge of early and accurate detection of glaucoma.