Energy-based binary segmentation of snow microtomographic images

P. Hagenmuller (1) , Guillaume Chambon (1) , B. Lesaffre (2) , F. Flin (2) , N. Calonne (2) , Mohamed Naaim (1)

1 UR ETGR (ETNA) - Erosion torrentielle neige et avalanches
2 Météo-France

X-ray micro-tomography has become an essential tool to investigate the mechanical and physical properties of snow, which are tied to its microstructure. A crucial step in the 3D image processing is the binary segmentation of the grayscale X-ray attenuation coefficient image to a binary ice/pore image. In the snow community, this step is usually based on global thresholding conducted with smoothing filters. In practice, this standard segmentation method presents drawbacks and often requires time-consuming manual post-processing. We use a segmentation method based on the minimization of an energy function. The energy definition formally expresses the segmentation criteria and clearly defines the effective resolution of the output image. Moreover, the global optimization through graph-cuts is particularly robust. We applied this method to different snow images and we successfully compared the results to standard segmentation techniques. Finally, the segmentation sensibility to the smoothness parameter is analyzed.

Mots clés: IMAGERIE; PROPRIETE MECANIQUE DE LA NEIGE; PROPRIETE PHYSIQUE DE LA NEIGE; SEGMENTATION D'IMAGE; MICRO TOMOGRAPHY; ENERGY BASED; GRAPH CUT; MICROSTRUCTURE; SEGMENTATION; SNOW

Domaines: Sciences de l'environnement

P. Hagenmuller, Guillaume Chambon, B. Lesaffre, F. Flin, N. Calonne, et al.. Energy-based binary segmentation of snow microtomographic images. 1st International Conference on Tomography of Materials and Structures (ICTMS), Jul 2013, Ghent, Belgium. 4 p. ⟨hal-00863099⟩