Our regional images are composed of many evaluations about the various kinds of elements or activities within different regions. In quantitative terms, the extraction of regional images is difficult. In this paper, we focus on a phenomenon in which a part of regional image is reflected in the names of buildings. In the first instance, a model based on the random utility theory is constructed to describe the spatial distribution of buildings names. Secondly, the proposed model is calibrated using actual data from the city of Tokyo (Setagaya Ward), and the effects of such regional elements or activities on regional images are then estimated. Finally, values for the regional images are quantitatively estimated and their spatial distribution is represented on a map.
