ABSTRACT

Map generalization is a process of hierarchically reorganizing features whereby the global shape of the original datasets can be transferred in different scales. We propose a stroke and centrality-based method to hierarchically extract the skeleton structures from buildings aiming to support generalization. Firstly, the strokes are generated from refined proximity graph network. Next, by regarding the strokes as dual graph, three centrality indices are calculated for each stroke whereby an integrated factor is created to measure the importance level of the strokes. Finally, the hierarchical skeleton structures are extracted based on the stroke importance levels through different selection ratios. By classifying the buildings into different categories, different generalization operators are selected considering their characteristics. The experimental results demonstrate that the extracted hierarchical skeleton structures can represent the global shape of the entire region. Through this support, the global and local patterns of the original buildings can be both preserved.

摘要

地图概化是一个分层重组特征的过程，原始数据集的全局形状可以在不同的比例下传输。我们提出了一种基于笔画和中心性的方法来分层地从建筑物中提取骨架结构，以支持泛化。首先，笔画是从细化的邻近图网络生成的。接下来，通过将笔画视为对偶图，为每个笔画计算三个中心性指数，从而创建一个综合因子来衡量笔画的重要性水平。最后，通过不同的选择比例，根据笔画重要性级别提取层次骨架结构。通过将建筑物分为不同的类别，根据其特点选择不同的泛化算子。实验结果表明，提取的层次骨架结构可以代表整个区域的全局形状。通过这种支持，可以同时保留原始建筑的全局和局部模式。