A diffusion process for wavelet-transform-based image denoising and its application to document image binarization

Lixin Fan, Liying Fan, Chew Lim TanICIP 2003 · OpenReview

Abstract

To distinguish image features from high level noises, we propose to exploit the spatial correlations between wavelet coefficients by replacing the thresholding process with a diffusion process. The proposed smoothing method is more robust to noise when applied to highly corrupted images.
BibTeX
@inproceedings{fan2003diffusion,
  author = {Lixin Fan and Liying Fan and Chew Lim Tan},
  title = {{A diffusion process for wavelet-transform-based image denoising and its application to document image binarization}},
  booktitle = {ICIP},
  year = {2003},
  url = {https://openreview.net/forum?id=zICU9Tcub5},
}

Loading related content...

Figure