An Edge Dependent Weighted Filter for Video Deinterlacing

Hao Zhang, Rong Meng-tian, Tao LiuIEICE 2015 · DOI

Abstract

In this letter, we propose a new intra-field deinterlacing algorithm based on an edge dependent weighted filter (EDWF). The proposed algorithm consists of three steps: 1) calculating the gradients of three directions (45°, 90°, and 135°) in the local working window; 2) achieving the weights of the neighboring pixels by exploiting the edge information in the pixel gradients; 3) interpolating the missing pixel using the proposed EDWF interpolator. Compared with existing deinterlacing methods on different images and video sequences, the proposed algorithm improves the peak signal-to-noise-ratio (PSNR) while achieving better subjective quality.
BibTeX
@inproceedings{zhang2015edge,
  author = {Hao Zhang and Rong Meng-tian and Tao Liu},
  title = {{An Edge Dependent Weighted Filter for Video Deinterlacing}},
  booktitle = {IEICE},
  year = {2015},
  doi = {10.1587/transfun.e98.a.788},
}

Loading related content...

Figure