Abstract
In this paper, a least distortion approach is proposed for halftone image watermarking. The impacts of distortion and tonality problems in halftoning are analyzed. An iterative linear gain model is developed to optimize perceptual quality of watermarking halftone images with efficient computation complexity O(1). An optimum linear gain for data hiding error diffusion is derived and mapped into a standard linear gain model, with the tonality evaluated using the average power spectral density. As compared with Fu and Au’s data hiding error diffusion method, our experiments show that our proposed linear gain model can achieve an improvement of between 6.5% to 12% using weighted signal-to-noise ratio (WSNR) and an improvement of between 11% to 23% measured by visual image fidelity (VIF).