The key idea of METAe is to systematically extract metadata from the layout as well as from structural and segmental elements of books simultaneously to the digitisation process. For hundreds of years books and journals have had a common layout and a common structure. Without understanding the language or the content of a book, readers are able to recognise title pages, tables of contents, appendices, graphs and pictures. But even more detailed elements such as headlines, page numbers, footnotes, typefaces, paragraphs or prose, verse and lyric elements can be detected just by the layout's characteristics. The objective of METAe will be to develop a software which is able to extract as much metadata as possible from the layout of a book and to transform it into XML
structured text. In addition to the text METAe will generate Dublin Core metadata and the digital facsimile of the document.