Ultra-low Resolution Character Recognition by Utilizing Multiple Frames

Bibliographic Information

Other Title
  • 動画像を用いた超低解像度文字認識(ポスターセッション,実世界指向,産業)

Search this article

Description

Improvement of character recognition technology brings us various character recognition applications for mobile camera. However, many low-resolution and poor-quality character images exist, due to the influence of environment or motion blur. Therefore, we develop a character recognition system for ultra-low resolution character images less than 20^*20 pixels. The proposed system consists of three phases of increased training data with a generative learning method, creating a deblurred high-resolution image with Wiener filter and image alignment, and recognition by efficient Mutual Subspace Method. We examined our system and confirmed the improvement of accuracy and efficiency of the recognition process.

Journal

  • Technical report of IEICE. PRMU

    Technical report of IEICE. PRMU 114 (454), 95-96, 2015-02-12

    The Institute of Electronics, Information and Communication Engineers

Keywords

Details 詳細情報について

  • CRID
    1571698602433138688
  • NII Article ID
    110010011727
  • NII Book ID
    AN10541106
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top