Wavelet feature based SVM and NAIVE BAYES classification of glaucomatous images using PCA and Gabor filter

The increase in intraocular pressure within the eye causes degradation of optic nerves which results in glaucoma. It is an eye disease in which no early symptoms will be detected until some vision loss has occurred. Therefore diagnosing of glaucoma is very essential to minimize the risk of vision loss. In this paper, the input retinal images are enhanced by using Principal Component Analysis and the blood vessels are removes by Gabor filter, morphological operation and thresholding techniques. Glaucomatous image classification is performed using texture features of an image. The texture features are obtained using 2-D discrete wavelet transform (DWT). The filters used in this paper are symlet3 (sym3) and bi-orthogonal (bio3.3, bio3.5). The extracted features are validated by support vector machine and Naive Bayes classifier. Finally the performance measures of the two classifiers are compared.