Multilayer Scene Similarity Assessment

Abstract

As we move increasingly towards multi-source data analysis, the assessment of similarity of complex, multilayer scenes is becoming increasingly important for spatial data mining. In this paper, we present a content-based approach for scene similarity assessment. The proposed approach is based on a graph-matching scheme that models linear feature networks (road network) as graphs and additional GIS information (e.g. buildings) as layer content. This allows us to combine diverse but co-located pieces of information (e.g. roads and buildings) in an integrated similarity assessment process. In the paper we present key theoretical concepts and provide experimental results to demonstrate the capability and robustness of the proposed approach.