Turning Blue Sound into Blue Noise

@inproceedings{tmb_icassp92,
    author  = "Bernard, T. ",
    title   = "Turning Blue Sound into Blue Noise",
  booktitle = "Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing",
    address = "San Francisco, California",
    pages   = "III:197-200",
    month   = mar,
    year    = 1992
}

Abstract
Because binary representation of images fit well VLSI constraints, the bilevel rendition of continuous tone pictures, refered to as the halftoning problem, is likely to play an important role in robot vision soon. In the present paper, we explain (in 1-D) and put in evidence (in 2-D) an important limitation of the most familiar high-quality technique, i.e. error diffusion. By extending earlier work of Gray, we show that the full propagation of the error on uniform images may generate a significant amount of low-frequency spectral energy, which corresponds to the appearance of undesirable textures and artifacts on the halftoned image. So, disagreeing with results presented by Ulichney, we conclude that error diffusion is a bad "blue noise" generator. To solve the difficulty, we relax the dc constraint through an optimization approach.

Paper (PostScript format compressed with gzip) -> paper.ps.gz (about 53K)

Back to global view of Thierry Bernard's publications