FM-HALFTONING WITH SURROUNDING-INFLUENCED QUANTIZATION AND LAZY PROPAGATION

摘要

The present invention refers to an image data processing method for processing input image data representing input color pixels of an input color image to be reproduced wherein the image data processing method uses an error diffusion process in order to produce output binary image data which represent an output binary image which represents a processed binary version of the input color image, the output binary image comprising output pixels having output binary color values; process pixels are determined by the error diffusion process, which process pixels constitute a process image and which respectively have a process color value, wherein the error diffusion process uses the process pixels in order to process the input color pixels step by step in order to stepwise determine the output binary values of the output pixels by quantization; wherein the process pixel determined from that one of the input color pixels which is processed in a current one of the steps is a current process pixel, characterised in that the following is performed in the current step: a) a current binary color value is set for a current output pixel which corresponds to the current process pixel; b) a plurality of reproduction color values of particular ones of the output pixels are respectively calculated based on output binary color values determined in previous steps and based on the current binary color value; c) an error indicator is determined based on a plurality of provisional quantization errors, said provisional quantization errors are calculated based on the differences between the reproduction color values of the particular output pixels and the process color values of the process pixels corresponding to the particular output pixels; wherein the steps a), b) and c) are repeated for different current binary color values; and, for quantization of the current pixel, that one of the different current binary color values is selected as output binary color value for the current output pixel for which the error indicator indicates the smallest error.