Fusion Graphs: Merging Properties and Watersheds

Abstract Region merging methods consist of improving an initial segmentation by merging some pairs of neighboring regions. In this paper, we consider a segmentation as a set of connected regions, separated by a frontier. If the frontier set cannot be reduced without merging some regions then we call it a cleft, or binary watershed. In a general graph framework, merging two regions is not straightforward. We define four classes of graphs for which we prove, thanks to the notion of cleft, that some of the difficulties for defining merging procedures are avoided. Our main result is that one of these classes is the class of graphs in which any cleft is thin. None of the usual adjacency relations on ℤ2 and ℤ3 allows a satisfying definition of merging. We introduce the perfect fusion grid on ℤn, a regular graph in which merging two neighboring regions can always be performed by removing from the frontier set all the points adjacent to both regions.

[1]  Gilles Bertrand,et al.  A Boolean characterization of three-dimensional simple points , 1996, Pattern Recognition Letters.

[2]  W. T. Tutte Graph Theory , 1984 .

[3]  Gilles Bertrand,et al.  Topological gray-scale watershed transformation , 1997, Optics & Photonics.

[4]  Gilles Bertrand,et al.  Fusion Graphs, Region Merging and Watersheds , 2006, DGCI.

[5]  Serge Beucher,et al.  Use of watersheds in contour detection , 1979 .

[6]  S. Beucher,et al.  Morphological segmentation , 1990, J. Vis. Commun. Image Represent..

[7]  C. Berge Graphes et hypergraphes , 1970 .

[8]  Gilles Bertrand,et al.  A new characterization of three-dimensional simple points , 1994, Pattern Recognition Letters.

[9]  Azriel Rosenfeld,et al.  Digital Picture Processing , 1976 .

[10]  Michael L. Baird Structural Pattern Recognition , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Theodosios Pavlidis Advanced Segmentation Techniques , 1977 .

[12]  Gilles Bertrand,et al.  Grayscale Watersheds on Perfect Fusion Graphs , 2006, IWCIA.

[13]  Azriel Rosenfeld,et al.  Digital Picture Processing, Volume 1 , 1982 .

[14]  Theodosios Pavlidis,et al.  Structural pattern recognition , 1977 .

[15]  Gilles Bertrand,et al.  Watersheds, mosaics, and the emergence paradigm , 2005, Discret. Appl. Math..

[16]  Gilles Bertrand,et al.  Weighted fusion graphs: Merging properties and watersheds , 2008, Discret. Appl. Math..

[17]  Gilles Bertrand,et al.  Enhanced computation method of topological smoothing on shared memory parallel machines , 2005, Journal of Mathematical Imaging and Vision.

[18]  Gilles Bertrand,et al.  Watershed Cuts: Minimum Spanning Forests and the Drop of Water Principle , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  James C. Alexander,et al.  The Boundary Count of Digital Pictures , 1971, JACM.

[20]  Gilles Bertrand,et al.  Quasi-Linear Algorithms for the Topological Watershed , 2005, Journal of Mathematical Imaging and Vision.

[21]  Azriel Rosenfeld,et al.  Digital topology: Introduction and survey , 1989, Comput. Vis. Graph. Image Process..