Cognitively Plausible Representations for the Alignment of Sketch and Geo-referenced Maps

Jan Sahib , Schwering Angela , Schultz Carl , and Chipofya Malumbo

Abstract

In many geo-spatial applications, freehand sketch maps are consid-ered as an intuitive way to collect user-generated spatial informa-tion. The task of automatically mapping information from such hand-drawn sketch maps to geo-referenced maps is known as the align-ment task. Researchers have proposed various qualitative represen-tations to capture distorted and generalized spatial information insketch maps, however, thus far the effectiveness of these representa-tions has not been evaluated in the context of an alignment task. Thispaper empirically evaluates a set of cognitively plausible representa-tions for alignment using real sketch maps collected from two differ-ent study areas with the corresponding geo-referenced maps. Firstly,the representations are evaluated in a single-aspect alignment ap-proach by demonstrating the alignment of maps for each individualsketch aspect. Secondly, representations are evaluated across multiplesketch aspects using more than one representation in the alignmenttask. The evaluations demonstrated the suitability of the chosen rep-resentation for aligning user-generated content with geo-referencedmaps in a real-world scenario.