Glaucoma Image Classification Using Entropy Feature and Maximum Likelihood Classifier
Anisha Rebinth, S. Mohan Kumar, T. Kumanan, G. Varaprasad
Abstract
Abstract In general, the nerve that links the eye to the brain is affected because of high eye pressure. The most common kind of glaucoma sometimes has no other symptoms than a gradual loss of vision. In this study, the Glaucoma Image Classification (GIC) is made by using different entropy features and Maximum Likelihood Classifier (MLC). Initially, the input fundus images are decomposed by using rankles transform, then the entropy features like sample entropy, Shannon entropy and approximate entropy are used to extract features. Finally, MLC is applied for classification. The GIC scheme’s function produces the classification accuracy of 96 % by using Shannon entropy feature and MLC.
Topics & Concepts
Pattern recognition (psychology)Artificial intelligenceGlaucomaEntropy (arrow of time)Principle of maximum entropyContextual image classificationClassifier (UML)Computer scienceMathematicsClassification schemeMachine learningImage (mathematics)PhysicsMedicineOphthalmologyQuantum mechanicsRetinal Imaging and AnalysisGlaucoma and retinal disordersDigital Imaging for Blood Diseases