计算机科学 ›› 2013, Vol. 40 ›› Issue (Z6): 192-195.

• 模式识别 • 上一篇    下一篇

基于分割线的漫画帧识别与自动排序

焦利敏,何中市,李佳   

  1. 重庆大学计算机学院 重庆400030;重庆大学计算机学院 重庆400030;重庆大学计算机学院 重庆400030
  • 出版日期:2018-11-16 发布日期:2018-11-16

Frame Recognition and Sort of Comic Pages Based on Split Line

JIAO Li-min,HE Zhong-shi and LI Jia   

  • Online:2018-11-16 Published:2018-11-16

摘要: 分析了漫画构成文法,提出一种漫画结构定义法;分析了漫画帧排布特点和分治思想的关联性,提出基于二叉树结构的漫画帧迭代识别排序法;分析了漫画分割线特点,提出一种基于分割比的分割线选择算法。本方法可以对漫画帧进行自动识别、抽取与排序,帮助减少手动分割的工作量。分割后的漫画帧可以放在屏幕较小的手机或PDA上阅读,使漫画阅读方便、简单、生动。实验证明该方法对大多数漫画帧的识别分割和故事情节的先后顺序排列具有较好的适应性与实时性 。

关键词: 漫画帧识别,分割线选择,漫画文法,直线检测,霍夫变换

Abstract: After analyzing comic grammar,this paper presented one kind of definition for comic structure.According to the relevance between features of frames layout and Divide-and-Conquer,iterative segmentation algorithm based on binary tree was proposed.After analyzing comic split line,split line choosing algorithm based on split ratio was proposed.The proposed method could decompose one comic page into multiple comic frames and sort them in order automatically,so it can reduce the amount of manual segmentation work.Comic frames after division can be placed on the smaller screen of cell phone screen or PDA,one by one in order to allow simple and vivid reading even through screen with very limited size.Experiments show that for most comics this method can identify each frame and then arranged them in chronological order according to the story and that it has better performance in both accuracy and real time.

Key words: Frame recognition,Split line choosing,Comic grammar,Line detection,Hough transform

[1] Yamada T,Watanabe T.Extraction of Person Objects from Japanese.Four-Scenes Comics [J].IEIC Technical Report,1999,99(47):152-159
[2] Hoashi K,Ono C,Ishii D,et al.Automatic Preview Generation of Comic Episodes for Digitized Comic Search [C]∥Proceedings of the 19th ACM international conference on Multimedia.2011:1489-1492
[3] Kurlander D,Skelly T,Salesin D.Comic Chat [C]∥Procee-dings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques.1996:225-236
[4] Yamada M,Budiarto R,Miyazaki S.Comic image decomposition for reading comics on cellular phones [J].IEICE transactions on information and systems,2004,87(6):1370-1376
[5] Tanaka T T,Shoji K,Toyama F,et al.Layout analysis of tree-structured scene frames in comic images [C]∥Proceedings of IJCAI.2007:2885-2890
[6] Chan C,Leung H,Komura T.Automatic panel extraction of color comic images [J].Advances in Multimedia Information Processing-PCM 2007,2007,4810:775-784
[7] Arai K,Herman T.Method for automatic e-comic scene frame extraction for reading comic on mobile devices[C]∥Information Technology:New Generations(ITNG),2010Seventh International Conference on.IEEE,2010:370-375
[8] In Y,Oie T,Higuchi M,et al.Fast frame decomposition and sorting by contour tracing for mobile phone comic images [J].International journal of systems applications,engineering and development.2011,5(2):216-223
[9] Nobuyuki O.A threshold selection method from gray-level histograms [J].IEEE transactions on systems,1979,9(1):62-66
[10] Fernandes L A F,Oliveira M M.Real-time line detection th-rough an improved Hough transform voting scheme[J].Pattern Recognition,2008,41(1):299-314
[11] 郭斯,霍文娟,唐求,等.结合Hough变化与改进最小二乘法的直线检测[J].计算机科学,2012,9(4):196-200
[12] Illingworth J,Kittler J.A survey of Hough transform [J].Computer Vision Graphics Image Processing,1988,44(1):87-116

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!