Visual fidelity of bilevel halftones generated by the error diffusion filter optimized for the blue noise masks in all tone levels

In this paper a set of the optimum error diffusion filters is presented for the bilevel halftoning of grayscale images and the visual fidelity of the bilevel halftones generated by the error diffusion filters is investigated. The error diffusion filter is optimized with respect to the ideal blue noise mask (BNM) corresponding to a single tone level. The number of the optimum filters is hence equal to that of different tones. The filter coefficients are optimized by the minimization of the squared error norm between the frequency response of the filter and the Fourier spectrum of the BNM. During the process of optimization, the binary pattern power spectrum manipulation algorithm (BIPPSMA) is applied. Visual fidelity of the bilevel halftones generated by the error diffusion filters is investigated in terms of a weighted signal-to-noise ratio (WSNR) and Fourier power spectra. Experimental results have demonstrated that the proposed filter set generates satisfactory bilevel halftones of grayscale images.