Edge Point Linking by Means of Global and Local Schemes

This chapter presents an efficient technique for linking edge points in order to generate a closed contour representation. The original intensity image, as well as its corresponding edge map, are assumed to be given as input to the algorithm (i.e., an edge map is previously computed by some of the classical edge detector algorithms). The proposed technique consists of two stages. The first stage computes an initial representation by connecting edge points according to a global measure. It relies on the use of graph theory. Spurious edge points are removed by a morphological filter. The second stage finally generates closed contours, linking unconnected edges, by using a local cost function. Experimental results with different intensity images are presented.