In this article, we evaluate the qualities of images obtained by the multi-level error diffusion method and propose objective parameters for evaluation of image quality. The results of the subjective evaluation experiments show that the image quality is improved as the number of output levels increase, but the improvement is saturated at 8 levels in 400 dpi resolution and 24 levels in 200 dpi. The objective parameters proposed in this study are rms (root mean square) granularity and SNR (signal to noise ratio). The results of the subjective evaluation show that these parameters are useful to evaluate the quality of the images obtained by the multi-level error diffusion method.
Kimiyoshi Miyata, Masayuki Saito, "An Evaluation Method for the Images Obtained by Multi-Level Error Diffusion Technique" in Journal of Imaging Science and Technology, 1998, pp 115 - 120, https://doi.org/10.2352/J.ImagingSci.Technol.1998.42.2.art00004