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 document images.
Citation:
Lixin Fan, Liying Fan, Chew Lim Tan, "Wavelet Diffusion for Document Image Denoising," icdar, vol. 2, pp.1188, Seventh International Conference on Document Analysis and Recognition (ICDAR'03) - Volume 2, 2003