Advanced Search
Volume 36 Issue 3
Apr.  2014
Turn off MathJax
Article Contents
Zhao Xiao-Ming, Zhao Yuan-Mei, Zhou Xiao-Yuan, Yuan Sheng-Chun. A Novel Color Anti-aliasing Method Related to Displays[J]. Journal of Electronics & Information Technology, 2014, 36(3): 721-727. doi: 10.3724/SP.J.1146.2013.00636
Citation: Zhao Xiao-Ming, Zhao Yuan-Mei, Zhou Xiao-Yuan, Yuan Sheng-Chun. A Novel Color Anti-aliasing Method Related to Displays[J]. Journal of Electronics & Information Technology, 2014, 36(3): 721-727. doi: 10.3724/SP.J.1146.2013.00636

A Novel Color Anti-aliasing Method Related to Displays

doi: 10.3724/SP.J.1146.2013.00636 cstr: 32379.14.SP.J.1146.2013.00636
  • Received Date: 2013-05-09
  • Rev Recd Date: 2013-10-16
  • Publish Date: 2014-03-19
  • Considering the color aliasing caused by down-sampling in display systems, a novel color anti-aliasing filtering method based on the subpixel arrangements of displays is proposed. Due to the shape and the area of color aliasing are closely related with the diversity of arrangements, the Nyquist limits of different primary colors within various arrangements are obtained in frequency domain by reciprocal lattice theory, and the anti-aliasing filters applied to different primary colors are designed. To suppress color aliasing efficiently while keeping more original image details, the images are reconstructed and displayed through subpixel sampling after preprocessed by them. Analysis and experiments show that when the clarity of display images is measured by the PSNR at the luminance component, the method proposed in this paper can increase the PSNR nearly 30% than 5-tap filtering. At the same time, the method is simple and easy to implement.

  • loading
  • 加载中

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Article Metrics

    Article views (2229) PDF downloads(820) Cited by()
    Proportional views
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return