Detail of Publication

Text Language	Japanese
Authors	Masakazu IWAMURA, Tomohiko TSUJI, Koichi KISE
Title	Real-Time Case-Based Recognition of Camera-Captured Characters
Journal	IEICE Technical Report
Vol.	109
No.	418
Presentation number	PRMU2009-222
Pages	pp.87-92
Reviewed or not	Not reviewed
Month & Year	February 2010
Abstract	In this paper, in the recognition problem of camera-captured characters, we present how to realize quickand precise recognition using a lot of case examples stored in a database. To be more precise, we introduce our fastand robust approximate nearest neighbor search method, originally designed for specic object recognition, to ourreal-time camera-based character recognition method. In the experiments using a database storing 100 fonts, weconrmed that the proposed method achieved high performance in both recognition rate and processing time.

• Note
  PRMUの原稿に誤りがあったため、訂正版を置いてあります。下記が大きな修正点です．
  • 図12(a)
    従来手法の認識率が間違っていて，本来の値より12%くらい低くなっていた．そのため，提案手法で上昇した認識率は約20%から約8%になった．
  • 3節の2段落目
    提案した3つの方策のうち，1番目のものの閾値の設定方法が間違っていた．
  • 図9
    図中の「生成された二値ベクトル」が5本になっていたが，正しくは3本だった．
• Following files are available.
1. PRMU201002_masa_rev.pdf (4.3MB)
2. PRMU201002_masa_rev.ppt (6MB)
3. PRMU201002_masa_rev.pptx (2.2MB)
4. paper_rev.pdf (1.7MB)
• Entry for BibTeX

  @InCollection{IWAMURA2010,
    author =	{Masakazu IWAMURA and Tomohiko TSUJI and Koichi KISE},
    title =	{Real-Time Case-Based Recognition of Camera-Captured Characters},
    booktitle =	{IEICE Technical Report},
    year =	2010,
    month =	feb,
    volume =	{109},
    number =	{418},
    presenID =	{PRMU2009-222},
    pages =	{87--92}
  }
  
  
  

Back to list