Claims
- 1. A method of remapping colors in an original document, preparatory to use in a system having an output system gamut different than originally applied, including:for a target pixel in a set of pixels defining at least a portion of an image, storing a set of neighboring pixels including said target pixel, wherein said target pixel and neighboring pixels form an N×M set of pixels in said image; analyzing said set of neighboring pixels, determining at least one cluster for pixels therein; determining at least one cluster metric for each said cluster; determining at least one key color within each said cluster in said set of pixels; using said at least one cluster metric, mapping said at least one key color to a color inside said output system gamut; from said at least one remapped key color, determining at least one mapping vector representing the change thereof; and using said at least one mapping vector to remap target pixel.
- 2. A method as defined in claim 1, wherein N=M=10.
- 3. A method as defined in claim 1, wherein said cluster metrics include a metric measuring cluster variance.
- 4. The method as defined in claim 3, wherein said key color is remapped to a new key color in said output system gamut, using variance measurements to assure that any cluster surrounding said key color is substantially included within said output system gamut.
- 5. The method as defined in claim 4, wherein greater values of said cluster variance measurement require remapping said key color a greater color distance beyond a boundary of said output color gamut.
- 6. A method as defined in claim 1, wherein said cluster metrics include a metric measuring cluster size.
- 7. A method as defined in claim 1, wherein said cluster metrics include a metric measuring the cluster mean.
- 8. The method as defined in claim 1, wherein said at least one key color is determined as an average value for any cluster in said set of neighboring pixels and said target pixel.
- 9. The method as defined in claim 1, wherein said key color is remapped to a new key color in said output system gamut, using said cluster metric measurements to assure that any cluster surrounding said key color is substantially included within said output system gamut.
- 10. The method as defined in claim 1, wherein subsequent to remapping said target pixel, said target pixel is compared with a boundary of said output system gamut to assure that said target pixel is there within.
- 11. The method as defined in claim 1, wherein said at least one key color number two or more.
- 12. The method as defined in claim 11, wherein subsequent to remapping said at least two key colors, luminance values thereof are compared to prior luminance values, and adjusted to assure that the luminance relationships between the said at least two key colors have been approximately maintained during remapping.
- 13. The method as defined in claim 1, wherein said target pixel is remapped to a color in said output gamut by an amount determined by a distance-weighted average of mapping vectors.
- 14. A method of remapping colors in an original document, preparatory to use in a system having an output system gamut different than originally applied, including:for a target pixel in a set of pixels defining at least a portion of an image, storing a set of neighboring pixels including said target pixel, wherein said target pixel and neighboring pixels form an N×M set of pixels in said image; identifying a color context within said set of pixels; using said color context, mapping said at least one key color to a color inside said output system gamut; from said remapped key color, determining an amount of change to be applied to said target pixel; and using said amount of change to remap said target pixel.
- 15. A device for remapping colors in an original document, preparatory to use in a system having an output system gamut different than originally applied, including:an image input device, receiving images defined in terms of pixels having colors of an input gamut; an image processing system, including a gamut remapping function, remapping input gamut defined pixels to output system gamut defined pixels; said image processing system including a computational device operating in accordance with a program providing that provides the following processing for each pixel in the received image: for a target pixel in a set of pixels defining at least a portion of an image, storing a set of neighboring pixels including said target pixel, wherein said target pixel and neighboring pixels form an N×M set of pixels in said image; analyzing said set of neighboring pixels, determining at least one cluster metric for clusters therein; determining at least one key color within said set; using said at least one cluster metric, mapping said at least one key color to a color inside said output system gamut; from said remapped key color, determining a mapping vector representing the change thereof; and using said mapping vector to remap target pixel; a document output, availing said remapped target pixels to a output device for display, print, or storage.
CROSS REFERENCE
Cross-reference is made to copending patent application Ser. No. 09/435,308, filed Nov. 5, 1999, entitled, “Gamut Mapping Preserving Local Luminance Differences” by Balasubramanian et al. and patent application Ser. No. 09/435,256, filed Nov. 5, 1999 entitled, “Gamut Mapping Preserving Local Luminance Differences with Adaptive Spatial Filtering”, by R. Balasubramanian et al. These applications are also incorporated by reference for their teachings.
US Referenced Citations (6)
Non-Patent Literature Citations (3)
Entry |
T.Morimoto et al, “Objects' Color Matchings Based on Image Clustering”, Japan Hardcopy '98, p. 371ff.* |
R. Rolleston, “Using Shepard's Interpolation to Build Color Transformation Tables”, Proceedings of IS&T/SID's Color Imaging Conference, 1994, pp. 74-77. |
John McCann: “Lessons Learned from Mondrians Applied to Real Images and Color Gamuts”, Proc. IS&T/SID seventh Color Imaging Conference, Nov. 1999, XP002188517, p. 7, column 2. |