Culture Knowledge Base

The Ground Truth Guidelines have grown out of the DFG-funded initiative for Optical Character Recognition Development. They address the creation of ground truth, i. e. reference data for the development and evaluation of OCR systems.

Specifications for the digitization of two-dimensional materials such as manuscripts, books, newspapers, and magazines, but also photographs, paintings, and graphic sheets. They are technically oriented and refer exclusively to the image quality and metadata of the primary file.

The "Masterplan Zeitungsdigitalisierung" summarizes the experiences and results of the DFG project "Digitalisierung historischer Zeitungen" and includes a summary of the most important recommendations.

The European Impact Centre of Competence in Digitisation has published this best practice guide to OCR. The guide provides an introduction to OCR and is more technical in nature, a bit older, but thorough.

This document presents professional methods for creating the most faithful digital reproductions possible of physical two-dimensional collection items such as texts, plans, maps, graphics, or photographs.Iit also focuses on documenting image-related information in metadata.

The DFG Practical Guidelines formulate standards and provide guidance on organisational, methodological and technical issues in the context of object digitisation and indexing. This document is an updated version of the Practical Guidelines published by the DFG in 2016.