Generalized statistical smoothing to the problem of inverse-halftoning for error diffusion

We improve the performance of the statistical smoothing to the problem of inverse-halftoning for error diffusion by introducing the general parameter scheduling. Then, using the numerical simulation for several gray-level images, we estimate the performance based on the mean square error, the histogram of the gray-level and the gray-levels on the segment across the edges of the original image. The simulations clarify that the performance of the statistical smoothing is improved by tuning the parameters of our model.