Segmentation of Natural Images Using Scale-Space Representation with Multi-Scale Edge Supervised Hierarchical Linking
O. Divorra Escoda, Pierre Vandergheynst · 2001
In general purpose computer vision systems, non-supervised image analysis is mandatory in order to achieve an automatic operation. In this paper a different approach to image segmentation for natural scenes is presented. Scale-Space representation is used to extract the structure from meaningful objects in the image. A hierarchical decomposition of the image is performed from the iso-intensity paths. The Scale-Space stack is generated using isotropic diffusion on the basis of linear Scale-Space theory.