"Scalable Methods of Text and Structure Recognition for the Full-Text Digitization of Historical Prints" Part 2: Layout Analysis

“历史印刷品全文数字化的文本和结构识别的可扩展方法”第 2 部分：布局分析

• 批准号: 394346204
• 负责人: Professor Dr. Andreas Dengel
• 金额: --
• 依托单位: Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI)
• 依托单位国家: 德国
• 项目类别: Research data and software (Scientific Library Services and Information Systems)
• 财政年份: 2018
• 资助国家: 德国
• 起止时间: 2017-12-31 至 2018-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/394346204?language=en
• 关键词: Scalable; Methods; Text; Structure; Recognition

The project “Skalierbare Verfahren der Text- und Strukturerkennung für die Volltextdigitalisierung historischer Drucke” has the goal of developing a complete OCR-Workflow for a high quality mass digitization of historical prints from the 16th-18th century. For each step in the workflow innovative methods should be made available as tools. Module 2: Layouterkennung ist next to OCR itself the most important step. It improves the OCR results directly, but also improves the general understanding of the digitized document by providing insights to the layout and relations between the document components. For each optimization step there are a wide variety of algorithms available, however not all of them are suitable to the specific challenges of this projects. On the basis of prior experience and work, the DFKI plans the identification, development and integration of suitable methods.

“Skalierbare Verfahren der Text- und Strukturerkennung für die Volltextdigitalisierung historischer Drucke”项目的目标是开发一个完整的 OCR 工作流程，用于对 16 世纪至 18 世纪的历史印刷品进行高质量的大规模数字化，工作流程中的每一步都采用创新方法。模块 2：Layouterkennung 是仅次于 OCR 本身的最重要的工具。它直接改进了 OCR 结果，而且还通过提供对文档组件之间的布局和关系的洞察来提高对数字化文档的总体理解。对于每个优化步骤，都有多种可用的算法，但并非全部。适合该项目的具体挑战 根据先前的经验和工作，DFKI 计划确定、开发和整合合适的方法。