A method for estimating morphological parameters of vessels in fundus images based on curve visibility matrix

We discuss an approach to fundus image analysis that enables one to modify the diagnostic vessel parameters in such a way that the morphological peculiarities of tree-like structures are accounted for. A feature estimation method based on processing the geometric characteristics of central vessel line lobes is described. An approach relying upon partitioning the curves into lobes with use of the algorithm of visibility matrix construction is discussed. The approach developed makes it possible to construct a primary feature space that can be used for constructing morphological features invariant to various types of geometric distortions.