Salta al contenuto principale
Passa alla visualizzazione normale.

GIOSUE' LO BOSCO

A genetic algorithm for image segmentation

Abstract

The paper describes a new algorithm for image segmentation. It is based on a genetic approach that allow us to consider the segmentation problem as a global optimization problem (GOP). For this purpose a fitness function, based on the similarity between images, has been defined. The similarity is function of both the intensity and spatial position of pixels. Preliminary result, obtained using real images, show a good performance of the segmentation algorithm.