Skip to content

This project employs Optical Character Recognition (OCR) to digitize historical records from the Qing manufacturing office.

License

Notifications You must be signed in to change notification settings

UranusSeven/qing_bureau_of_construction

Repository files navigation

清宫造办处电子档案 Digital Records of Qing Manufacturing Office

这个项目用 OCR 技术识别清宫造办处档案, 并用全文搜索技术为学者提供方便的搜索能力.

This project employs Optical Character Recognition (OCR) to digitize historical records from the Qing manufacturing office. By utilizing full-text indexing and searching technology on the OCR results, it offers historians an efficient and convenient method to search through the digital records.

本项目原理如下图所示:

Here's an overview of how the project functions:

Untitled Diagram

其中, OCR 的工作通过 古籍酷 提供的 OCR API 完成. 十分感谢古籍酷提供的 OCR 识别 API, 大大加快了本项目的开发进度.

Thanks gj.cool for their powerful OCR API, which greatly accelerated the development of this project.

内容 Content

由于清宫造办处档案数量十分庞大, 我计划分批完成电子档案的构建.

26/01/2023 更新:

  • 添加了 49卷 前 50 页内容

27/01/2023 更新:

  • 添加了 49卷 50 页至 96 页内容

28/01/2023 更新:

  • 添加了 49卷 96 页至 168 页内容

29/01/2023 更新:

  • 添加了 49卷 剩余部分
  • 添加了 48卷 前 127 页内容

30/01/2023 更新:

  • 添加了 48卷 剩余部分
  • 添加了 44卷47卷 全部内容
  • 添加了 43卷 前 209 页内容

11/02/2023 更新:

  • 添加了 43卷 剩余部分
  • 添加了 39卷42卷 全部内容
  • 添加了 50卷55卷 部分内容

12/02/2023 更新:

  • 添加了 55卷 的全部內容

使用 Usage

克隆本项目 Clone the Project

SSH:

git clone git@github.com:UranusSeven/qing_bureau_of_construction.git

HTTPS:

git clone https://github.com/UranusSeven/qing_bureau_of_construction.git

安装依赖 Install Dependencies

为了避免依赖版本冲突,强烈建议创建新的虚拟环境. 之后可以通过 pip 安装依赖.

To avoid conflicts, it is highly recommended to create a new virtual environment before installing the dependencies. You can then install the dependencies with pip.

pip install -r requirements.txt

运行

执行 streamlit run app.py,浏览器会自动打开搜索界面。此时系统会自动检测索引是否存在,第一次运行时系统将会自动构建索引。

Run streamlit run app.py. You should see a new pop-up on your browser. When you run the system for the fisrt time, the system will automatically build the index.

image

输入关键字后按回车,系统将展示所有匹配结果。如果结果中 XX 卷 XX 頁X半部分 是超链接,单击可以打开 PDF 文件的对应页。(目前仅在 macOS + Chrome 环境下验证过,Chrome 需要安装 Enable local file links 插件)

You can then input the keywords and press ENTER and the system will show you all the matches. If you want to see the original PDF page for context, just click the hyperlink in the results. (Works for macOS + Chrome. Chrome plugin Enable local file links is required)

image

高级搜索 Advanced Search

目前支持输入以空格分隔的多个关键字,系统将会展示同时包含这些关键字的结果。

You can input space separated keywords for records that contains all of these keywords.

About

This project employs Optical Character Recognition (OCR) to digitize historical records from the Qing manufacturing office.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages