Classification-Driven Stochastic Watershed: Application to Multispectral Segmentation

Guillaume Noyel (1) , Jesus Angulo (1) , Dominique Jeulin (1)

1 CMM - Centre de Morphologie Mathématique

The aim of this paper is to present a general methodology based on multispectral mathematical morphology in order to segment multispectral images. The methods consists in computing a probability density function pdf of contours conditioned by a spectral classification. The pdf is conditioned through regionalized random balls markers thanks to a new algorithm. Therefore the pdf contains spatial and spectral information. Finally, the pdf is segmented by a watershed with seeds (i.e., markers) coming from the classification. Consequently, a complete method, based on a classification-driven stochastic watershed is introduced. This approach requires a unique and robust parameter: the number of classes which is the same for similar images. Moreover, an efficient way to select factor axes, of Factor Correspondence Analysis (FCA), based on signal-to-noise ratio on factor pixels is presented.

Mots clés: remote sensing; classification-driven stochastic watershed; marginal probability density function; regionalized random balls; regionalized random markers; multispectral images; Stochastic Watershed; Machine learning

Domaines: Traitement des images [eess.IV]; Traitement du signal et de l'image [eess.SP]; Traitement d'image

Guillaume Noyel, Jesus Angulo, Dominique Jeulin. Classification-Driven Stochastic Watershed: Application to Multispectral Segmentation. 4th European Conference on Colour in Graphics, Imaging, and Vision and 10th International Symposium on Multispectral Colour Science (CGIV 2008/MCS'08), Jun 2008, Terrassa - Barcelona, Spain. pp.471-476. ⟨hal-00830667⟩