Artificial neural networks in retinal image analysis

Glaucoma disease detection from retinal images using classifiers like Least Square-Support Vector Machine classifier, Random forest, Dual Sequential Minimal Optimization classifier, Naive bayes classifier and Artificial neural networks. The textual features obtained from retinal images are used for this classification. Energy distributions over wavelet sub bands provide these features. The proposed system is using discrete wavelet transform to extract different wavelet features obtained from the three filters symlets (sym3), daubechies (db3) and bi-orthogonal (bio3.3, bio3.5, and bio3.7) wavelet filters. The energy signatures obtained from 2-D discrete wavelet transform is used for classifying and detecting glaucomatous and normal retinal images.