along a space-filling curve over the image. Unfortunately, this method tends to blur image details if the cluster size is Ordered dither (C. N. Judice, J. F. Jarvis, and W. H. Ninke, in Proceedings of the S.I.D. 15, 4 (Fourth Quarter), 1974, reasonably large. This problem is solved in the stochastic pp. 161-169.) is an efficient and widely used halftoning techscreening dithering method [6] by performing an adaptive nique. One disadvantage with this method is that it produces variation of the cluster size. Like its predecessor [5], howblurred images. Recently, a space-filling curve ordered dithever, this method is computationally expensive and not suitering approach (Y. Zhang, submitted for publication.) has been able for parallel processing.
proposed that improves the ordered dither in image quality by
Recently, Zhang [8] proposed a space-filling curve orgenerating clustered dot patterns along a space-filling curve dered dither that can significantly improve the space-filling over the image. In this method the cluster size is fixed in the curve clustering method [5] in both speed and image qualprocess of halftoning. This paper presents a new ordered dithity. In this method, the pixels of the image are divided into ering method, called adaptive ordered dither, that does halftonclasses by subdividing the space-filling curve over the image ing by using a space-filling curve to perform an adaptive variainto segments as the space diffusion [9] does. Then, like tion of the cluster size. Experimental results demonstrate that in ordered dither , the class numbers associated with the new method can significantly improve the space-filling curve ordered dither in revealing image details. ยฉ 1997 Academic Press the pixels of the image are used as the dithering thresholds of these pixels. However, this method is still limited in producing pleasing images due to a difficulty with choosing an appropriate cluster size (see Section 2.2).