The article presents findings from the interdisciplinary research-project “Signs of the Metropolises: Visual multilingualism in the Ruhr-Area/Germany” (funded by MERCUR), which investigates the occurrence, regional distribution, function, production and perception of visual multilingualism in representative neighbourhoods of the cities Duisburg, Essen, Bochum and Dortmund. The focus of this paper will be on the structure and features of an online data base – the main corpus of the project – and its application opportunities for the sociolinguistic and linguistic landscape research. The image database encompasses about 25,595 geo-coded digital photographs. The photographs are linked to a map of the Ruhr Area to show the distribution and density of visual multilingualism. The image database also provides metadata for the following categories according to Scollon and Scollon (2003) and Backhaus (2007): choice of language/variety, type of discourse (e.g. commercial, regulatory, transgressive, commemorative), type of name (e.g. institution, shop, gastronomy, toponym), information management (e.g. complete, partial, extended), appearance, typography (e.g. Antiqua, Grotesque, Fraktur type, handwritten), and size. This tagging system allows complex search strategies to analyse sociolinguistic and geographical aspects of visual multilingualism.