关键词: 拓扑关系, 紧凑图像表达, 自适应, 矢量路网

Abstract:

The 2D image representation of road network aims to establish a road-to-image transformation relationship, which is valuable for many transportation prediction problems such as road traffic flow prediction. However, current studies to represent road network can be affected by the topological relationship loss problem and the spatial resolution problem. Thus, a 2D image representation method for road network is proposed. The method can be adapted to road networks with different spatial structures and can derive a compact 2D image from the road network by mapping the individual road segments into the corresponding image pixels in terms of maintaining their topological relationships in a large extent. Experiments are conducted to evaluate the performance of the method in hundreds of urban road networks at home and abroad, and the results suggest the effectiveness and superiority of the proposed method by comparing with random coding and sequential coding.

Key words: spatial topological relationship, compact image representation, adaptive, vector based road network