• DocumentCode
    1227328
  • Title

    An Intelligent Subtitle Detection Model for Locating Television Commercials

  • Author

    Huang, Yo-Ping ; Hsu, Liang-Wei ; Sandnes, Frode-Eika

  • Author_Institution
    Dept. of Comput. Sci. & Eng., Tatung Univ., Taipei
  • Volume
    37
  • Issue
    2
  • fYear
    2007
  • fDate
    4/1/2007 12:00:00 AM
  • Firstpage
    485
  • Lastpage
    492
  • Abstract
    A strategy for locating television (TV) commercials in TV programs is proposed. Based on the observation that most TV commercials do not have subtitles, the first stage exploits six subtitle constraints and an adaptive neurofuzzy inference system model to determine whether a frame contains a subtitle or not. The second stage involves locating the mark-in/mark-out points using a genetic algorithm. An interactive user interface allows users to efficiently identify and fine-tune the exact boundaries separating the commercials from the program content. Furthermore, erroneous boundaries are manually corrected. Experimental results show that the precision rate and recall rates exceed 90%
  • Keywords
    adaptive systems; fuzzy neural nets; genetic algorithms; interactive television; telecommunication computing; user interfaces; TV program; adaptive neurofuzzy inference system; genetic algorithm; intelligent subtitle detection model; interactive user interface; television commercial; Adaptive systems; Advertising; Databases; Genetic algorithms; Gunshot detection systems; Humans; Multimedia communication; TV broadcasting; User interfaces; Visual system; Adaptive neurofuzzy inference system (ANFIS); TV commercial locating; genetic algorithm (GA); subtitle detection; television (TV) commercial detection; Advertising as Topic; Algorithms; Artificial Intelligence; Commerce; Image Interpretation, Computer-Assisted; Natural Language Processing; Pattern Recognition, Automated; Television; Video Recording; Vocabulary, Controlled;
  • fLanguage
    English
  • Journal_Title
    Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1083-4419
  • Type

    jour

  • DOI
    10.1109/TSMCB.2006.883428
  • Filename
    4126288