Image Segmentation Using a Genetic Algorithm
Vitoantonio Bevilacqua, Giuseppe Mastronardi · 2003
In this paper we present a genetic algorithm-based optimisation technique for an automatic selecting of the thresholds in image segmentation, considering in a combined way, the parameters of the segmentation and the parameters of the pre-processing and post-processing operators. The design of a database based on the most important image’s features to generate an accurate initial population for all the needed parameters is described. The obtained performances of the designed system are then shown qualitatively. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.