Text retrieval for Japanese historical documents by image generation

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Digitization of historical documents is growing rapidly. Text retrieval is a vital technology to facilitate the use of historical document images because of the large amount of data. In this paper, we propose a method for retrieving keywords in Japanese historical documents with text query. The proposed method automatically generates an image of the query text and retrieves regions in documents similar to the generated image by feature matching. We exploit a technique of deep learning to generate an image close to texts in Japanese historical document images. Furthermore, we use con-volutional neural network to extract features robust to appearance variation of texts in documents, such as shade and shape of texts. We conducted the text retrieval experiments on the public dataset of Japanese historical documents in the Edo era. The experimental results show the effectiveness of the proposed method.

Original languageEnglish
Title of host publicationHIP 2017 - Proceedings of the 2017 Workshop on Historical Document Imaging and Processing
PublisherAssociation for Computing Machinery
Pages19-24
Number of pages6
ISBN (Electronic)9781450353908
DOIs
Publication statusPublished - 2017 Nov 10
Event4th International Workshop on Historical Document Imaging and Processing, HIP� 2017 - Kyoto, Japan
Duration: 2017 Nov 102017 Nov 11

Publication series

NameACM International Conference Proceeding Series

Other

Other4th International Workshop on Historical Document Imaging and Processing, HIP� 2017
CountryJapan
CityKyoto
Period17/11/1017/11/11

Keywords

  • Document retrieval
  • Generative model
  • Historical document
  • Image processing
  • Neural networks

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Text retrieval for Japanese historical documents by image generation'. Together they form a unique fingerprint.

Cite this