A perception-based color recommendation algorithm for hierarchical regions

The visualization of hierarchical and nested spatial regions remains a challenge to cartographers. Despite progress in computer algorithms for visualizing general hierarchical data, mapping spatial hierarchical regions, especially with static, noninteractive means, still requires considerable manual efforts. This paper proposes a two-step algorithm that can automatically recommend perception-based colors and help reveal the hierarchical structure embedded in nested regions. It first systematically sorts regions according to their contiguity and containment relations at multiple hierarchical levels. Then, perception-based colors are generated using the order of regions with the goal of maximizing differentiability between top-level regions while retaining the perceived uniformity of the bottom-level regions. With the coloring scheme recommended by this algorithm, the metric color differences among regions mathematically reflect their hierarchical positions and spatial relations. The resultant colors, therefore, can potentially help map-readers perceive the spatially constrained hierarchy structure built in nested regions.