Partial Difference Equations over Graphs: Morphological Processing of Arbitrary Discrete Data
Mathematical Morphology (MM) offers a wide range of operators to address various image processing problems. These processing can be defined in terms of algebraic set or as partial differential equations (PDEs). In this paper, a novel approach is formalized as a framework of partial difference equations (PdEs) on weighted graphs. We introduce and analyze morphological operators in local and nonlocal configurations. Our framework recovers classical local algebraic and PDEs-based morphological methods in image processing context; generalizes them for nonlocal configurations and extends them to the treatment of any arbitrary discrete data that can be represented by a graph. It leads to considering a new field of application of MM processing: the case of high-dimensional multivariate unorganized data.
KeywordsWeight Function Texture Image Weighted Graph Mathematical Morphology Graph Topology
Unable to display preview. Download preview PDF.
- 2.Brockett, R., Maragos, P.: Evolution equations for continuous-scale morphology. In: IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 3, pp. 125–128 (1992)Google Scholar
- 4.Maragos, P.: PDEs for morphology scale-spaces and eikonal applications. In: Bovik, A. (ed.) The Image and Video Processing Handbook, 2nd edn., pp. 587–612. Elsevier Academic Press, Amsterdam (2004)Google Scholar
- 7.Gilboa, G., Osher, S.: Nonlocal operators with applications to image processing. Report 07-23, UCLA, Los Angeles (July 2007)Google Scholar
- 8.Peyré, G.: Manifold models for signals and images. Technical report, CEREMADE, Université Paris Dauphine (2007)Google Scholar