Rapidly Creating Structure-Aware Halftoning with Improved Error Diffusion

Showing the details of edge and texture structure clearly is an important aspect of modern halftoning technique. This paper proposes a simple and effective error diffusion halftoning algorithm, and with keeping structure-aware on show. The method is based on the well-known Ostromoukhov algorithm. According with local area gray distribution, we calculate the local visual information of texture feedback (ITF) around each pixel and then the ITF is used to adjust the fixed threshold. The results of experiment show that achieved halftone images have clarity and natural texture structure. It not only represents ideal blue noise spectrum, but also shows a better performance in evaluation index of peak signal to noise ratio (PSNR), universal image quality index (UIQI) and structural similarity index measure (SSIM).