Objectives  The central time-space maps deform the map space to visualize the one-to-many time distance, which can intuitively reflect the distance changes due to the influence of a traffic system. The core of the central time-space map construction is to calculate the non-control points' locations based on the control points' displacements. Displacement is a basic operation in cartographic generalization and many approaches have been developed.

  Methods  We model the central time-space map construction as a problem of displacement in cartographic generalization, use the relative neighborhood graph to express the proximity relations between the points, and the initial forces on the points are computed based on the control points' displacements.And the Snake model is applied iteratively based on the built graph to obtain the new locations of all points, post operations are also applied to avoid topology errors in the iterative process.

  Results and Conclusions  Compared to the existing approaches, the proposed method can reduce the topology errors and improve the shape similarity for the deformed boundaries, and can better avoid topology errors and maintain the local morphologies.