Distance evaluation in pattern matching based on frontier topological graph
We focus to an application of control of printing quality in industrial environment. We try to value the quality in the sense of the readability of a text printed in disturbed conditions. We propose in this paper, topological comparison criteria between two superposed patterns, based on a graph representation.
The graph is built according to the regions and their adjacencies in the resulting image of a maximal superposition of two patterns: each edge of the graph is not only joining two regions, but also note a frontier between two regions. The study of this graph highlight creation or loss of connectivity, appearance or disappearance of holes or meaningful part between the compared patterns. This topological study allows us to propose an interesting distance evaluation function between the two patterns to be compared.
KeywordsDistance Evaluation Function Graph Representation Pattern Matching
- [AAF95]Ehoud Ahronovitz, Jean-Pierre Aubert, and Christophe Fiorio. The star-topology: a topology for image analysis. In 5 th Discrete Geometry for Computer Imagery, Proceedings, pages 107–116. Groupe GDR PRC/AMI du CNRS, september 1995.Google Scholar
- [dF97]Gilles d'Andréa and Christophe Fiorio. Maximal superposition of grids and an application. In 7 th International Workshop, DGCI'97, number 1347 in Lecture Notes in Computer Sciences, pages 63–71, Montpellier, France, December 1997.Google Scholar
- [Fio95]Christophe Fiorio. Approche interpixel en analyse d'images: une topologie et des algorithmes de segmentation. Thèse de doctorat, Université Montpellier II, 24 novembre 1995.Google Scholar
- [Fio96]Christophe Fiorio. A topologically consistent representation for image analysis: the frontiers topological graph. In 6 th International Workshop, DGCI'96, number 1176 in Lecture Notes in Computer Sciences, pages 151–162, Lyon, France, November 1996.Google Scholar
- [GW87]Rafael C. Gonzalez and Paul Wintz. Digital image processing. Addison-Wesley, second edition, 1987.Google Scholar
- [Pav77]Theo Pavlidis. Structural Pattern Recognition. Springer-Verlag, New York, 1977.Google Scholar
- [Vos92]George Vosselman. Relational Matching. Lecture Notes in Computer Science. Springer-Verlag, Berlin Heidelberg Germany, 1992.Google Scholar