Key points are not available for this paper at this time.
An automatic document entry system is described that identifies the type of document and extracts textual information, such as titles or authors, from semi-formatted document images. The system registers documents, offers easy retrieval of documents used in a daily workflow analyzes the layout structure of documents by using document specific models, and assumes that each type of document is known in advance. In this paper we focus on a method for identifying the type of document.
Kochi et al. (Fri,) studied this question.