Glaucoma diagnosis using support vector machine

This article discusses some techniques used for glaucoma diagnosis. in this paper a supervised learning is proposed and compared with other existing techniques for the detection of glaucoma. For classification support vector machines principle is used. Nonlinear transformation of the support vectors is used and classification can be done. The retinal fundus images of the databases are taken for the glaucoma detection. The proposed classifier will take these images and formation of the presence/ absence of glaucoma by calculating the CDR value in the respected image. The features of the images are extracted by using PCA (principal component analysis). The accuracy, predictability of the proposed classifier is compared with the techniques which includes fused method and Support vector machine by morphological based image classification.