Abstract / Description of output
A road network is cartographically drawn in varying levels of detail depending on the resolution or scale of the output graphic. In automated generalization, the challenge is in deriving generalized forms of a road network, appropriate for the intended target scale and to achieve this with minimum intervention from the user. This paper presents a method for the detection and simplification of road junctions as part of that process. Road junctions within the network are identified using a combination of spatial clustering and graph theory. The junctions are simplified using a combination of contractions and restrictions of the graph. Consideration is given to ways in which attribute and cartometric information can be used both to modify the behavior of the algorithm, and to influence the choice of other generalization algorithms. This algorithm is considered to be part of a growing number of generalization algorithms that can be used to derive generalized products from single detailed database. The success and limitations are discussed and future developments are proposed.
Keywords / Materials (for Non-textual outputs)
- Automated map generalization
- Cluster analysis
- Graph theory
- Junction simplification