BDIMFs Based Features Fusion and Classification for Glaucoma Prediction

There are various disorders in the humen eye. Glaucoma is one of the most critical eye disorders. Eye is a precious part of the human body. It develops in the eye with very less speed and damages all internal optic nerves. Outdated glaucoma prediction methods are more costly and slow. Various research works have been reported using multiresolution approaches and various features. A better work for glaucoma prediction using bi-dimensional intrinsic mode functions (BDIMFs) based features fusion and classification from fundus image is proposed. Adaptive BDIMFs are obtained from preprocessed color components. BDIMFs based features extraction and their fusion is the key idea of this work. BDIMFs based features are then normalized and classified with support vector machines (SVM). The pooled feature accuracy reached 97.7 with 10-fold cross-validation (FCV). Experimental analysis clearly shows the superiority of the proposed method over the traditional method.