A DYNAMIC ROI BASED GLAUCOMA DETECTION AND REGION ESTIMATION TECHNIQUE

Glaucoma is a visual issue caused because of expanded fluid pressure in the optic nerve. It harms the optic nerve and causes the loss of eye sight. The former methods checking strategies are filtering laser polarimetry, Heidelberg retinal tomography etc. These techniques are costly, require expert technician to operate and have problem in segmentation and boundary scaling. So there is a need to diagnose glaucoma with a minimal effort. Hence we propose a new strategy called the Region of Interest (ROI) based glaucoma detection and region estimation technique, in which first ROI is selected and then the Empirical Wavelet Transform (EWT) is applied. The EWT is utilized to break image into multiple positive and negative scenarios. The correntropy features are obtained from these EWT segments, these features are ranked based on the threshold value selection algorithm. Then these features are used for the classification of ordinary and glaucoma affected image by utilizing the LS-SVM classifier.