Colour quantisation through dithering techniques

Most colour image printing and display devices do not have the capacity to deal with true colours. To overcome this limitation, these devices need to apply a quantisation method to reduce the number of colours present in the image. The research of this paper focuses on the study of a special set of quantisation methods known as dithering techniques. Several quantisation techniques already exist that produce good quality results which however are still not perfect. A new method is proposed in this contribution based on the same principles as those pertaining to the sigma-delta modulators used in analogue-to-digital conversion but extended to the two-dimensional image space. The results of this research show that the quality of the quantised image is improved because the quantisation noise is filtered more efficiently by the noise-shaping filter. Furthermore, the complexity of the method remains comparable with existing algorithms, such as the error diffusion method.