Work Package 7 develops an intelligent and collaborative system for the extraction of structured information from images of historical documents related to companies’ financial and economic activities. The system focuses on printed serial sources related to listed companies such as stock exchange yearbooks and price lists. In the deliverable D7.1, we provided general software libraries, which can be used to build different prototypes of document recognition and understanding systems adapted to different kinds of documents. This milestone M7.1 is composed of the first version of two recognition systems: one for yearbook information extraction, and one for price list data extraction. Those systems have been applied to several corpora: the German Yearbook 1913-1914 (Handbuch der deutschen Aktiengesellschaften), the French Desfossés Yearbook 1962, the official price lists for Brussels 1912 and Paris 1961-1962, which are part of the document samples dataset validated by the Steering Committee. This document samples dataset is made of three yearbooks, three stock price lists, with three different languages, on three time periods: before WWI, interwar and post WWII. The two remaining corpora are the Spanish yearbook 1929-1930 and the official price lists for Madrid 1934.