HOG feature based SVM classification of glaucomatous fundus image with extraction of blood vessels

Glaucoma at the later stages causes eye blindness, it is essential for early detection to minimize the risks and warn patients who might eventually lose their vision. The study gives an assessment on Glaucoma detection using the Histogram of Oriented Gradients (HOG) Feature extraction along with SVM classification of retinal fundus image with the extraction of blood vessels using Gabor filter. Retinal fundus images are reduced in their dimension. Then the blood vessels are extracted by using Gabor filter, morphological operations and thresholding. This improves the accuracy. Textural features within the images are used for accurate and efficient glaucoma classification. The image is then given to a HOG feature descriptor. The feature sets (statistical properties) are then extracted from the HOG Features that are used for Support Vector Machine classification of glaucomatous images.