Efficient Matching of Dynamically Changing Graphs

Bunke, Horst; Glauser, Thomas; Tran, Thanh Ha

doi:10.1142/9789812797896_0012

Search citation statements

Order By: Relevance

Paper Sections

Select...

Introduction1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

1995

Publication Types

Select...

Book1

Relationship

Self Cite1

Independent0

Authors

Journals

Cited by 1 publication

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Numerous image analysis applications of graph matching have been described in the literature. These include character classification [15], schematic diagram and 2-D shape analysis [14], 3-D object recognition [8,26], stereo vision [10], dynamic scene analysis [3] and muscle tissue classification [20].…”

Section: Introductionmentioning

confidence: 99%

Efficient attributed graph matching and its application to image analysis

Bunke

Messmer

1995

Image Analysis and Processing

Self Cite

View full text Add to dashboard Cite

Graphs are a very powerful data structure for many tasks in image analysis. If both known models and unknown objects are represented by graphs, the detection or recognition problem becomes a problem of graph matching. In this paper, we first review different methods for graph matching. Then we introduce a new family of exact and errortolerant graph matching algorithms that have a number of interesting properties. The algorithms are particularly efficient if there is a large number of model graphs to be matched with an unknown input graph. Moreover, they allow the incremental updating of a database of known models. This property supports the application of graph matching in a machine learning context. As an example, we show a 2-D shape recognition system based on graph matching that is able to learn new shapes.

show abstract