Page Under Construction

The aim of this project is to estimate the coordinates of the center of the macula region in fundus images. This method enables a fully automatic analysis and diagnosis of the macula (for example calculating pigment density using spectral filtered images).

A prototype (unclean code, preliminary) to test the first ideas and provide preliminary results was implemented in matlab. It compares multiple variants of different methods using results of the vessel segmentation and optic disk localization methods or gold standard data to fit a model, and refine its results by calculating local features in a region of interest. For the preliminary tests the HRF database is used, which has to be copied in the directories: images (input images), gs_vessels (gold standard segmentation), mask (ROI images)