Assisting Application for Analytical Transcription of Historical Documents based on Full-text Search System of Document Images.

我们开发了一个用于历史文档分析转录的辅助应用程序，该应用程序基于全文搜索技术，用于无法识别单个字符的图像扫描文档。该方法不依赖于语言和字体的差异，因为它使用了一种新的基于字形统计特征的伪编码方案。在本项目中，我们开发了一个历史文献辅助应用和后端搜索服务器作为服务，采用HMM相关反馈方法和字体稳健伪编码方法，支持草书手稿抄写员，加速了历史材料的分析。

We have developed an assisting application for analytical transcription of historical documents based on full-text search technique for image-scanned documents that does not recognize individual characters. The proposed method works independently of differences in language and font because it uses a new pseudo-coding scheme based on the statistical features of character shapes. In this project, we have developed an assisting application for historical document and back-end search server as a service, relevance feedback method with HMM and font robust pseudo-coding method.The system support transcribers of cursive manuscripts and accelerate the analysis of historical materials.