Information Extraction from Geographical Overview Maps Roman Pawlikowski, Krzysztof Ociepa, Urszula Markowska-Kaczmar, and Pawel B. Myszkowski Wroclaw University of Technology, Poland urszula.markowska-kaczmar@pwr.wroc.pl Abstract. The paper presents a method of information extraction from overview maps. The idea is based on recognizing text located on the map and on finding locations corresponding to the extracted text labels using the GeoNames ontology. The method consists of three phases. The first one performs map image processing in order to recognize text labels. The next phase verifies these labels and marks them as being sure or unsure locations. In the third phase the map is interpreted based on the locations found. The second and third phases make use of the ontology. The prelim- inary results are promising for further development of the method. Keywords: information extraction, overview map, map interpretation. 1 Introduction Maps have always been an important means of presenting useful information. There are many types of maps including physical, topographic, administrative, political, satellite and economical ones. Regardless of their purpose, they all retain the basic goal of presenting information with regard to location in space. Generally, a map is a form of graphical representation, an image similar to a chart, an illustration, or a photography. In fact, maps often contain elements of these different image categories or become variations of them. With the development of Geographical Information Systems (GIS), map acquisition and recognition have become hot top- ics. This paper is concerned with acquiring useful information through map pro- cessing. The method of knowledge extraction from maps described here is one of the tasks developed in a greater project which aim is to build a system capable of solving general problems given a user query. The user will communicate with the system using natural language. The system will answer the user query on the basis of knowledge it acquired from a huge corpus of text documents, many of them illus- trated with images. Knowledge extracted from text based on natural language pro- cessing methods will be supported by information extracted from images included in these documents. In this paper we focused on the method performing informa- tion extraction from maps. This paper is organized as follows. The next section presents motivations and formulates the research problem. Section 3 describes the proposed method of information extraction from maps. Section 4 presents some preliminary results of experiments. The last section discusses the research per- formed and shows future work in the development of the method. N.-T. Nguyen et al. (Eds.): ICCCI 2012, Part I, LNAI 7653, pp. 94–103, 2012. c  Springer-Verlag Berlin Heidelberg 2012