Integration of generalisation operators for vegetation maps based on bio-geographical principles

Two serious problems that often arise in map generalization are "what should be aggregated" and "when to aggregate". This very much depends on the knowledge of the theme to be mapped. In the case of vegetation mapping at different scales, some inherent rules on vegetation, such as their biological characteristics and spatial distribution, must be incorporated in addition to existing generalisation algorithms, so that automated generalisation can be achieved. This paper describes the development of such rules based on biogeographical principles for the implementation of existing generalisation operators and, in particular, the integration of these operators.