A study on large-scale manga image classification method

IR (HANDLE) Open Access
  • Yamanaka, Issa
    Department of Electrical Engineering and Computer Science, School of Engineering, Kyushu University
  • Ito, Eisuke
    Research Institute for Information Technology, Kyushu University

Bibliographic Information

Other Title
  • 大規模マンガ画像判別手法の検討

Description

Pirate edition of Manga is a serious problem. Since manga is published in digital e-book and paper book, it is easy to scan manga images. So DRM (digital rights management) is not so effective to against piracy. In order to suppress pirated manga, we are researching a discriminator for manga titles, volume numbers, story numbers, and character names from manga images. If classification accuracy is sufficient and the processing speed is high, it is possible to realize manga image filter. And the filter will deter the use of pirated copies. We applied the previously proposed classification method and its improvement plans to relatively large manga images. In this paper, we report the new classification method and the classification results.

Journal

Keywords

Details 詳細情報について

  • CRID
    1050862545765474816
  • HANDLE
    2324/7170850
  • Text Lang
    ja
  • Article Type
    conference paper
  • Data Source
    • IRDB

Report a problem

Back to top