Image Halftoning by Error Diffusion: A Survey of Methods for Artifact Reduction

Grayscale error diffusion introduces nonlinear distortion (directional artifacts and false textures), linear distortion (sharpening), and additive noise. In color error diffusion what color to render is a major concern in addition to finding optimal dot patterns. This article presents a survey of key methods for artifact reduction in grayscale and color error diffusion. The linear gain model by Kite et al. replaces the thresholding quantizer with a scalar gain plus additive noise. They show that the sharpening is proportional to the scalar gain. Kite et al. derive the sharpness control parameter value in threshold modulation (Eschbach and Knox, 1991) to compensate linear distortion. False textures at mid-gray (Fan and Eschbach, 1994) are due to limit cycles, which can be broken up by using a deterministic bit flipping quantizer (Damera-Venkata and Evans, 2001). Several other variations on grayscale error diffusion have been proposed to reduce false textures in shadow and highlight regions, including green noise halftoning (Levien, 1993) and tone-dependent error diffusion (Li and Allebach, 2002). Color error diffusion ideally requires the quantization error to be diffused to frequencies and colors, to which the HVS is least sensitive. We review the following approaches: color plane separable (Kolpatzik and Bouman 1992) design; perceptual quantization (Shaked et al. 1996, Haneishi et al. 1996); green noise extensions (Lau et al. 2000); and matrix-valued error filters (Damera-Venkata and Evans, 2001)