Mimetic Finite Difference Methods for Restoration of Fundus Images for Automatic Glaucoma Detection

Abstract

Glaucoma is one of the leading causes of irreversible blindness in people over 40 years old. In [9] it was developed a computational tool for automatic glaucoma detection, which implements a novel method that has shown improvement in the accuracy of the detection compared to other classical methods. However, the method is sensitive to the quality of the acquired image, which is often contaminated by noise, and its quality can be poor. For this reason, automatic image restoration of the source images is needed to improve the quality of glaucoma detection. Partial differential equations to produce an image of much higher quality, enhance its sharpness, filter out the noise, extract shapes, etc. Here, we proposed the use of mimetic finite difference methods for the numerical solution of this kind of problems. The mimetic methods preserve the continuum properties of the mathematical operators often encountered in the image processing and analysis equations and ensuring better orders of convergence [5]. By ensuring these mathematical properties, the original structure of the source image is maintained, improving the diagnosis of the patient.