古旧地图的信息化

On the Informationization of Antique Maps in Historical Geography

近年人文社会科学研究中古旧地图的使用规模和数字化处理需求不断增加,古旧地图的信息化迫在眉睫。文章以“数字历史黄河”(DHYR)·图形资料库建设为例,介绍DHYR中图形史料的RDF编目方案和古旧地图地名信息自动化;通过清代基层水利单元“汛”“堡”重建黄河变迁事件研究案例,分析古旧地图以及古旧地图信息化方案在人文社会科学研究中的意义。RDF是用于描述知识图谱实体及其关系的数据模型,能减少历史地理信息化数据孤岛现象,在DHYR中设计针对古旧地图的编目方案和描述词表;采用U-Net架构模型进行古旧地图地名OCR识别,展现深度学习方法在古旧地图信息自动化提取中的能力,提高历史地名提取的效率和准确率。

Antique maps are of unique and irreplaceable significance in historical geography,and their informationization has been more and more urgent.DHYR,or Digital Historical Yellow River,has been devoting itself to the collection and exploitation of graphic materials related to Yellow River.This paper first gives an introduction to DHYR’s RDF cataloging scheme of graphic historical materials and the automatic extraction of geographical name information of antique maps.It then makes a case study on grassroots water conservancy units(i.e.,Xun汛or Bao堡),making clear the changes of Yellow River’s watercourse in the Qing Dynasty,so as to show the significance of antique maps as well as the informationization of antique maps in historical geography.In detail,RDF is a data model proposed by W3C for describing knowledge mapping entities and their relationships,and it is able to effectively eliminate or reduce the phenomenon of“data silos”which is common in informationization of historical geography.As for DHYR,RDF is used to design a cataloging scheme and a descriptive list for antique maps.Besides,the U-Net architecture model is used for the OCR of geographical names of antique maps,which shows that the“deep learning”method could be used to improve the efficiency and accuracy of the automatic extraction of historical geographical names.