Cellular Neural Networks for Object-oriented Segmentation
Giuseppe Grassi, Pietro Vecchio · 2006
By exploiting the cellular neural network paradigm, this paper present a new object-oriented segmentation algorithm that takes into account the hardware characteristics of the cellular neural networks universal machine. In particular, by using a rigorous model of the image contours, this paper focuses on the edge extraction phase as well as on the performance evaluations. Simulation results, carried out for Stefan, coast guard, car-phone and football video sequences, show the effectiveness of the approach developed herein