of Eye Fundus Images Using Vessels
Detection and Hough Transform
F. Zana* and J. C. Klein
Abstract—Image registration is a real challenge because physicians handle many images. Temporal registration is necessary
in order to follow the various steps of a disease, whereas multimodal registration allows us to improve the identification of
some lesions or to compare pieces of information gathered from different sources. This paper presents an algorithm for temporal and/or multimodal registration of retinal images based on point correspondence. As an example, the algorithm has been applied to the registration of fluorescein images (obtained after a fluorescein dye injection) with green images (green filter of a color image). The vascular tree is first detected in each type of images and bifurcation points are labeled with surrounding vessel orientations. An angle-based invariant is then computed in order to give
a probability for two points to match. Then a Bayesian Hough transform is used to sort the transformations with their respective likelihoods. A precise affine estimate is finally computed for most likely transformations. The best transformation is chosen for registration.
Index Terms— Hough transform, image registration, mathematical morphology, opthalmology, retina.
REGISTRATION of medical images is fundamental in
order to follow the evolution of some lesions over
the years or to compare images obtained under different
conditions. This is particularly true for retinal images of eye fundus from patients developing diabetes. Different steps of most retinal diseases can be characterized by various damages of blood vessels. Temporal registration of images is necessary in order to follow the evolution of the illness, but multimodal registration is also a great challenge in the scope of an
automatic diagnosis. For example, in the case...