Font Size: a A A

Digital halftoning using a blue-noise mask

Posted on:1992-07-16Degree:Ph.DType:Dissertation
University:University of RochesterCandidate:Mitsa, TheophanoFull Text:PDF
GTID:1478390017950047Subject:Engineering
Abstract/Summary:
Many printing devices and displays are bilevel, and, therefore, are not capable of reproducing gray-scale images. Gray-scale images are converted to binary images using halftone processes. The two most widely used halftone processes are ordered dither and error diffusion. Ordered dither is a computationally fast method, but displays periodic artifacts in the halftone image. On the other hand, error diffusion produces high quality halftones, but requires computationally intensive filtering operations.;In this dissertation, a novel digital halftone process is presented, where the halftoning is achieved by a pixelwise comparison of the gray-scale image to an array, the blue-noise mask. The blue-noise mask is a dispersed-dot halftone screen, which combines the speed of ordered dither with the high quality results of error diffusion. This mask is designed so that the halftone image has blue-noise (high-frequency) characteristics in the transform domain. In the image domain, blue-noise characteristics correspond to visually pleasing, isotropic, uncorrelated patterns, without low-frequency graininess. The blue-noise mask is constructed to have unique first-order and second-order properties. When thresholded at any level, for example at A% of the maximum level, exactly A out of every 100 pixels will have a value greater than the threshold value. In addition, the spatial distribution of black-versus-white pixels is arranged in such a manner as to form a blue-noise pattern.;A visual study is also described, where subjects rated halftone images with the same first-order, but different second-order statistics. The purpose of this study was to determine how variations in the second-order statistics of a binary pattern affect its visual appearance. The quality of images halftoned using the new algorithm, and also ordered dither and error diffusion techniques, was measured using psychovisual tests and quantitative criteria. The success of the quantitative criteria, as quality measures for halftones, was assessed by comparing their results with the results of the psychovisual tests. Psychovisual tests indicate that halftoning using a blue-noise mask outperforms ordered dither techniques, and gives comparable results with error diffusion techniques.
Keywords/Search Tags:Blue-noise mask, Using, Error diffusion, Ordered dither, Halftoning, Images, Results
Related items