Clustered halftoning with dot-to-dot error diffusion

Claim 1

An Image processing system for use in quantizing multibit image signals for use at an output device (141), each image signal representing an optical density of an original image at a discrete location therewithin, and having an original optical density selected from one of a set of 'c' original optical density levels that has a number of members larger than a desired output set of 'd' desired optical density levels, said image processing system including: an input image buffer (100), suitable for storing at least a portion of a scanned image defined by multibit image signals and to which image signals are directed from a source (10) thereof; buffer means (104,106,108) coupled to said input image buffer (100) for receiving said image signals therefrom; signal adder means (110,112,114) for adding image signals directed thereto from said buffer means and error term signals ε(n) directed thereto from an error buffer (120), to produce modified image signals; halftoning means (124) coupled to said signal adder means (110,112,114), converting each modified image signal in an m pixel x n pixel area of the image to output signals, said conversion process varying with each m x n area of the image based on a predetermined dot growth pattern; means (130) directing said output signals to an output buffer (140), for transfer to the output device (141); density determining means for determining density of the image over the m x n area; memory means, storing a lookup table (152) of empirical output device responses to device output signals for an m x n area; difference means (154) for determining the difference between the stored empirical output density and the density determined from said density determining means, as a halftoning error ε(c); and distributing means (158,190,120) for distributing said halftoning error on a weighted basis to a plurality of image signals forming halftone cells, said distributing means including the error buffer (120) coupled to said signal adder means (110,112,114).

Claim 7

A method of quantizing image signals in an image formed by a plurality of multibit image signals for use at an output device, each multibit image signal representing the magnitude of density of the image at a location therewithin and having an original optical density level selected from one of a set of 'c' original optical density levels that has a number of members larger than a desired output set of 'd' desired optical density levels required for the output device, including: receiving a halftone cell set of image signals having original optical density levels selected from the set of 'c' original optical density levels, said image signal set corresponding to a halftone cell; modifying each image signal in the halftone cell set with an single error term signal derived from processing at least one previous halftone cell, and obtained from an error buffer memory; determining an input density of an area corresponding to the halftone cell, from the modified image signals in the halftone cell set; determining from the modified image signals a representative set of output device signals, said determination process varying with each m x n area of the image in accordance with a predetermined dot growth pattern; determining an output density of the area corresponding to the halftone cell from output values of the image signal corresponding to the cell; using said determined output density as an address in a lookup table stored in memory, said table producing an empirical output density, indicative of the output device response to said output device signals; determining the difference between a density defined by said modified input signals and said empirical output density within the m x n area as halftoning error; and storing a portion of said halftone error, to be added to image signals in at least one halftone cell.

Claim 10

A method as claimed in any one of claims 1 to 9, wherein "d" equals two, and said density determination is performed by incrementing a count for each image signal in the m x n area defined at the second of the two levels.