Algorithms for Extracting Textual Characters in Color Video

Edward Wong, Minya Chen

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Abstract

    In this chapter, we present a new robust algorithm for extracting text in digitized color video. The algorithm first computes the maximum gradient difference to detect potential text line segments from horizontal scan lines of the video. Potential text line segments are then expanded or combined with potential text line segments from adjacent scan lines to form text blocks, which are then subject to filtering and refinement. Color information is then used to more precisely locate text pixels within the detected text blocks. The robustness of the algorithm is demonstrated by using a variety of color images digitized from broadcast television for testing. The algorithm performs well on JPEG images and on images corrupted with different types of noise. For video scenes with complex and highly textured backgrounds, we developed a technique to reduce false detections by utilizing multiframe edge information, thus increasing the precision of the algorithm.

    Original languageEnglish (US)
    Title of host publicationComputer-Aided Intelligent Recognition Techniques and Applications
    PublisherJohn Wiley & Sons, Ltd
    Pages33-49
    Number of pages17
    ISBN (Print)0470094141, 9780470094143
    DOIs
    StatePublished - Dec 20 2005

    Fingerprint

    Color
    Television
    Pixels
    Testing

    Keywords

    • Artifact filtering step
    • Extra scan lines
    • maximum gradient difference (MGD)
    • Multiframe edge information
    • New robust algorithm
    • New text extraction algorithm
    • Text block detection
    • Text line segment
    • Video databases

    ASJC Scopus subject areas

    • Engineering(all)

    Cite this

    Wong, E., & Chen, M. (2005). Algorithms for Extracting Textual Characters in Color Video. In Computer-Aided Intelligent Recognition Techniques and Applications (pp. 33-49). John Wiley & Sons, Ltd. https://doi.org/10.1002/0470094168.ch3

    Algorithms for Extracting Textual Characters in Color Video. / Wong, Edward; Chen, Minya.

    Computer-Aided Intelligent Recognition Techniques and Applications. John Wiley & Sons, Ltd, 2005. p. 33-49.

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Wong, E & Chen, M 2005, Algorithms for Extracting Textual Characters in Color Video. in Computer-Aided Intelligent Recognition Techniques and Applications. John Wiley & Sons, Ltd, pp. 33-49. https://doi.org/10.1002/0470094168.ch3
    Wong E, Chen M. Algorithms for Extracting Textual Characters in Color Video. In Computer-Aided Intelligent Recognition Techniques and Applications. John Wiley & Sons, Ltd. 2005. p. 33-49 https://doi.org/10.1002/0470094168.ch3
    Wong, Edward ; Chen, Minya. / Algorithms for Extracting Textual Characters in Color Video. Computer-Aided Intelligent Recognition Techniques and Applications. John Wiley & Sons, Ltd, 2005. pp. 33-49
    @inbook{68b3b7663c9345469fb347d048dbf38c,
    title = "Algorithms for Extracting Textual Characters in Color Video",
    abstract = "In this chapter, we present a new robust algorithm for extracting text in digitized color video. The algorithm first computes the maximum gradient difference to detect potential text line segments from horizontal scan lines of the video. Potential text line segments are then expanded or combined with potential text line segments from adjacent scan lines to form text blocks, which are then subject to filtering and refinement. Color information is then used to more precisely locate text pixels within the detected text blocks. The robustness of the algorithm is demonstrated by using a variety of color images digitized from broadcast television for testing. The algorithm performs well on JPEG images and on images corrupted with different types of noise. For video scenes with complex and highly textured backgrounds, we developed a technique to reduce false detections by utilizing multiframe edge information, thus increasing the precision of the algorithm.",
    keywords = "Artifact filtering step, Extra scan lines, maximum gradient difference (MGD), Multiframe edge information, New robust algorithm, New text extraction algorithm, Text block detection, Text line segment, Video databases",
    author = "Edward Wong and Minya Chen",
    year = "2005",
    month = "12",
    day = "20",
    doi = "10.1002/0470094168.ch3",
    language = "English (US)",
    isbn = "0470094141",
    pages = "33--49",
    booktitle = "Computer-Aided Intelligent Recognition Techniques and Applications",
    publisher = "John Wiley & Sons, Ltd",

    }

    TY - CHAP

    T1 - Algorithms for Extracting Textual Characters in Color Video

    AU - Wong, Edward

    AU - Chen, Minya

    PY - 2005/12/20

    Y1 - 2005/12/20

    N2 - In this chapter, we present a new robust algorithm for extracting text in digitized color video. The algorithm first computes the maximum gradient difference to detect potential text line segments from horizontal scan lines of the video. Potential text line segments are then expanded or combined with potential text line segments from adjacent scan lines to form text blocks, which are then subject to filtering and refinement. Color information is then used to more precisely locate text pixels within the detected text blocks. The robustness of the algorithm is demonstrated by using a variety of color images digitized from broadcast television for testing. The algorithm performs well on JPEG images and on images corrupted with different types of noise. For video scenes with complex and highly textured backgrounds, we developed a technique to reduce false detections by utilizing multiframe edge information, thus increasing the precision of the algorithm.

    AB - In this chapter, we present a new robust algorithm for extracting text in digitized color video. The algorithm first computes the maximum gradient difference to detect potential text line segments from horizontal scan lines of the video. Potential text line segments are then expanded or combined with potential text line segments from adjacent scan lines to form text blocks, which are then subject to filtering and refinement. Color information is then used to more precisely locate text pixels within the detected text blocks. The robustness of the algorithm is demonstrated by using a variety of color images digitized from broadcast television for testing. The algorithm performs well on JPEG images and on images corrupted with different types of noise. For video scenes with complex and highly textured backgrounds, we developed a technique to reduce false detections by utilizing multiframe edge information, thus increasing the precision of the algorithm.

    KW - Artifact filtering step

    KW - Extra scan lines

    KW - maximum gradient difference (MGD)

    KW - Multiframe edge information

    KW - New robust algorithm

    KW - New text extraction algorithm

    KW - Text block detection

    KW - Text line segment

    KW - Video databases

    UR - http://www.scopus.com/inward/record.url?scp=84889967015&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84889967015&partnerID=8YFLogxK

    U2 - 10.1002/0470094168.ch3

    DO - 10.1002/0470094168.ch3

    M3 - Chapter

    AN - SCOPUS:84889967015

    SN - 0470094141

    SN - 9780470094143

    SP - 33

    EP - 49

    BT - Computer-Aided Intelligent Recognition Techniques and Applications

    PB - John Wiley & Sons, Ltd

    ER -