2.3.2 Retinal vessel centerline detection

One of the very useful applications in image processing and in shape recognition is skeletonization. The skeletons represent interconnected lines at the center of an object. They do not only describe the shape but also some mathematical properties of objects, such as length or surface [33–35]. The skeletons of the blood vessels are produced by a thinning technique [35–38]. In this study, the skeleton makes it possible to represent in a compact manner the properties of the blood vessels of the human retina in its particular shape. Its purpose is to describe each vessel by a set of infinitely fine lines. To achieve this objective, we have used the thinning method; the basic idea of this method is to reduce an object iteratively. The algorithm of this method consists of two main steps: the first step is detecting the contour of the object, and in the second step, the contour points are deleted only if they do not change the object's topology.
