ABSTRACT
In this chapter, the applicability of quantitative graph theory methods in the context of image analysis is studied. Whereas graph models themselves have been used for image analysis for a long time (see the edited book by Lezoray and Grady [38] for an overview of methods), graph indices have to our knowledge not played a significant role in this area so far. The work presented here is to be understood as a first step in this direction. As such, it focuses on one specific problem field, texture analysis, where a statistical approach to information encoded in large collections of graphs can be expected to have great potential.
