Performance estimation of generalized statistical smoothing to inverse halftoning based on the MTF function of human eyes

  • Authors:

  • Yohei Saika;Kouki Sugimoto;Ken Okamoto

  • Affiliations:

  • Department of Electrical and Computer Engineering, Wakayama National College of, Technology, Wakayama, Japan;Department of Electrical and Computer Engineering, Wakayama National College of, Technology, Wakayama, Japan;Department of Advanced Engineering, Wakayama National College of Technology, Wakayama, Japan

  • Venue:
  • ICA3PP'10 Proceedings of the 10th international conference on Algorithms and Architectures for Parallel Processing - Volume Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

We construct a method of the generalized statistical smoothing (GSS) to the problem of inverse halftoning for a halftone image which is converted by the error diffusion method. Especially, we construct the present method so as to achieve the optimal performance on the basis of the mean square error (MSE) between original and restored images both of which are observed through the MTF function of human vision system. Using the numerical simulation for several 256-level standard images, we clarify that the optimal performance of the GSS is realized if we appropriately set the parameters controlling both edge enhancement procedure and generalized parameter scheduling. We also find the GSS restores the original image more accurately than other conventional filters, such as the average and Gaussian filters.