Most halftoning algorithms assume there is no interaction between neighboring dots or if there is, it is additive. Without accounting for dot-gain effect, the printed image will not have the appearance predicted by the halftoning algorithm. Thus, there is need to embed a printer model in the halftoning algorithm which can predict such deviations and develop a halftone accordingly. The direct binary search (DBS) algorithm employs a search heuristic to minimize the mean squared perceptually filtered error between the halftone and continuous-tone original images. We incorporate a measurementbased stochastic model for dot interactions of an electro-photographic printer within the iterative DBS binary halftoning algorithm. The stochastic model developed is based on microscopic absorptance and variance measurements. We present an efficient strategy to estimate the impact of 5×5 neighborhood pixels on the central pixel absorptance. By including the impact of 5×5 neighborhood pixels, the average relative error between the predicted tone and tone observed is reduced from around 21% to 4%. Also, the experimental results show that electrophotography-model based halftoning reduces the mottle and banding artifacts.

Date Published: 18 January 2010
PDF: 14 pages
Proc. SPIE 7528, Color Imaging XV: Displaying, Processing, Hardcopy, and Applications, 752810 (18 January 2010); doi: 10.1117/12.843968

Published in SPIE Proceedings Vol. 7528:
Color Imaging XV: Displaying, Processing, Hardcopy, and Applications
Reiner Eschbach; Gabriel G. Marcu; Shoji Tominaga; Alessandro Rizzi, Editor(s)