Morphological Segmentation on Learned Boundaries

Allan Hanbury (1) , Beatriz Marcotegui (2)

1 PRIPG - Pattern Recognition and Image Processing Group
2 CMM - Centre de Morphologie Mathématique

Colour information is usually not enough to segment natural complex scenes. Texture contains relevant information that segmentation approaches should consider. Martin et al. [Learning to detect natural image boundaries using local brightness, color, and texture cues, IEEE Transactions on Pattern Analysis and Machine Intelligence 26 (5) (2004) 530-549] proposed a particularly interesting colour-texture gradient. This gradient is not suitable for Watershed-based approaches because it contains gaps. In this paper, we propose a method based on the distance function to fill these gaps. Then, two hierarchical Watershed-based approaches, the Watershed using volume extinction values and the Waterfall, are used to segment natural complex scenes. Resulting segmentations are thoroughly evaluated and compared to segmentations produced by the Normalised Cuts algorithm using the Berkeley segmentation dataset and benchmark. Evaluations based on both the area overlap and boundary agreement with manual segmentations are performed.

Mots clés: Color; Digital image storage; Image processing; Image segmentation; Landforms; Optical properties; Textures; Normalised cuts; Segmentation evaluation; Volume extinction values; Waterfall; Watershed

Domaines: Traitement des images [eess.IV]

Allan Hanbury, Beatriz Marcotegui. Morphological Segmentation on Learned Boundaries. Image and Vision Computing, 2009, 27 (4), pp.480-488. ⟨10.1016/j.imavis.2008.06.012⟩. ⟨hal-00833285⟩