MGS2 : Optimisation multicritères de contours actifs par algorithmes génétiques

Content Provider	Semantic Scholar
Author	Cladel, Nicolas Séguier, Renaud
Copyright Year	2005
Abstract	We propose in this paper an evolution of our preceding work on the optimization multicriterion of active contours. This approach allows a global analysis of the image and an efficient use of several energy. Our new algorithm, MultiObjective Genetic Snakes 2 (MGS2) associates the multiobjective algorithm NSGA2 [4] the coding of active contours of the MGS [3]. The MGS2 exploit new energies of fastener to the image in order to ensure the convergence of contours towards a hollow object in disturbed environment starting from a random initialization. We carry out here the study of these energies and the analysis of the convergence of the algorithm through a series of tests on a basis of synthetic objects. We will also have results of the application of MGS2 at the labial reading.
File Format	PDF HTM / HTML
Alternate Webpage(s)	http://documents.irevues.inist.fr/bitstream/handle/2042/13965/A341_78626.pdf;sequence=1
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article