Multifractal Analysis of Human Retinal Vessels

In this work it is shown that vascular structures of the human retina represent geometrical multifractals, characterized by a hierarchy of exponents rather then a single fractal dimension. A number of retinal images from the STARE database (www.parl.clemson.edu/stare) are analyzed, corresponding to both normal and pathological states of the retina. In all studied cases a clearly multifractal behavior is observed, where capacity dimension is always found to be smaller then the information dimension, which is in turn always smaller then the correlation dimension, all the three being significantly lower then the DLA (Diffusion Limited Aggregation) fractal dimension. We also observe a tendency of images corresponding to the pathological states of the retina to have lower generalized dimensions and a shifted spectrum range, in comparison with the normal cases.