Method for color rendering using scanned halftone classification

• US 8,184,340 B2
• Filed: 06/17/2003
• Issued: 05/22/2012
• Est. Priority Date: 06/17/2003
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Loce, Robert P., Maltz, Martin S., Grzenda, Jay S.
• Primary Examiner(s): Pathak, Sudhanshu C
• Assistant Examiner(s): Vo, Quang N
• Application Number:
  US10/462,966
• Publication Number:
  US 20040257619A1
• Time in Patent Office: 3,262 Days
• Field of Search: 358/2.1, 358/534, 358/1.9, 358/518, 358/3.23, 358/3.07, 382/162, 382/173, 382/224
• US Class Current: 358/2.1
• CPC Class Codes: H04N 1/40075 Descreening, i.e. converting a halftone signal into a corresponding continuous-tone signal; Rescreening, i.e. combined descreening and halftoning

Abstract

What is disclosed is an image path that advantageously uses halftone classification to select appropriate mappings in gray-scale management and color management operations. The tags generated in the scanner help identify different classes of halftones. One is selected from several pixel-value mappings to provide proper compensation. That is, the one-dimensional and multi-dimensional pixel-value mappings within the color management module are selected based on halftone classification tags from the scanner. The tagging is either one bit that indicates “Low Frequency Halftone” and “Not LFHT”, or, more preferably, the tag is multi-bit indicating a frequency bin that contains the frequency of the input halftone. Additionally, the multi-bit tag can indicate particular halftone screen types, such as dot screens, line screens, stochastic screens or error diffusion. The pixel value mappings are typically implemented as Look-Up Tables (LUTs), and the LUTs of the present invention are optimized for the various halftone classes. Generation of the LUTs, or other pixel-value mappings, is non-trivial in that such mappings are normally applied to contone images and not to blurred halftoned images. The pixel-value mappings of the present invention are derived via an iterative process. Alternatively, a model for this modulated image type is used.

Claims

1. A method for obtaining a preferred color conversion implementation for rendering of an image containing halftoned data, comprising:

providing a plurality of pixel-value to color rendering value mappings, each pixel-value to color rendering value mapping corresponding to a distinct color conversion implementation for converting a pixel-value of the image to a pixel-value in a colorspace used for rendering the image, each color conversion implementation corresponding to a distinct halftone type;

processing, using a processor, the image and generating therefrom a plurality of tags denoting the frequencies of halftoned data of the image containing halftoned data;

identifying different classes of halftones from the tags to obtain a plurality of halftone classifications;

selecting one of the halftone classifications;

selecting one of the plurality of pixel-value to color rendering value mappings, based upon the selected halftone classification, to obtain the preferred color conversion implementation for rendering the image containing halftoned data; and

rendering, using a printer, the image containing halftoned data using the selected pixel-value to color rendering value mapping.

2. The method for obtaining a preferred color conversion implementation, as in claim 1, further comprising performing spatial analysis and/or signal processing on the image to determine regions containing input halftones.

3. The method for obtaining a preferred color conversion implementation, as in claim 1, wherein the pixel-value to color rendering value mappings are look-up tables, each look-up table corresponding to a distinct color conversion implementation.

4. The method for obtaining a preferred color conversion implementation, as in claim 1, further comprising generating a set of calibration curves from said halftone classifications.

5. The method for obtaining a preferred color conversion implementation, as in claim 1, wherein the pixel-value to color rendering value mappings are tone reproduction curves, each tone reproduction curve corresponding to a distinct color conversion implementation.

6. The method for obtaining a preferred color conversion implementation, as in claim 1, wherein the pixel-value to color rendering value mappings are multi-dimensional.

7. The method for obtaining a preferred color conversion implementation, as in claim 1, wherein the pixel-value to color rendering value mappings are specialized mathematical functions, each specialized mathematical function corresponding to a distinct color conversion implementation.

8. The method for obtaining a preferred color conversion implementation, as in claim 1, further comprising obtaining at least one mapping between colors observed and an output rendering of CMYK separations.

9. The method, as in claim 1, wherein the pixel-value to color rendering value mappings are matrices, each matrix corresponding to a distinct color conversion implementation.

10. The method, as in claim 1, wherein the pixel-value to color rendering value mappings are sigmoid functions, each sigmoid function corresponding to a distinct color conversion implementation.

11. The method, as in claim 1, further comprising using look-up tables and/or tone reproduction curves to determine a location of at least one boundary region in colorspace.

12. The method, as in claim 11, wherein an optimum boundary region in colorspace is a function of halftone frequency response.

13. The method, as in claim 11, wherein an initial estimate of any one colorspace boundary is preferred until a better estimate of that boundary location is known.

14. A system for obtaining a preferred color conversion implementation for rendering of an image containing halftoned data, comprising:

a memory including a plurality of pixel-value to color rendering value mappings, each pixel-value to color rendering value mapping corresponding to a distinct color conversion implementation for converting a pixel-value of the image to a pixel-value in a colorspace used for rendering the image, each color conversion implementation corresponding to a distinct halftone type; and

a processing unit;

said processing unit generating a plurality of tags denoting the frequencies of halftoned data of the image containing halftoned data;

said processing unit identifying different classes of halftones from said tags to obtain a plurality of halftone classifications;

said processing unit selecting one of the halftone classifications; and

said processing unit one of the plurality of pixel-value to color rendering value mappings, based upon the selected halftone classification, to obtain the preferred color conversion implementation for rendering the image containing halftoned data.

15. The system, as in claim 14, performing spatial analysis and/or signal processing on the image to determine regions containing input halftones.

16. The system, as in claim 14, wherein the pixel-value to color rendering value mappings are look-up tables, each look-up table corresponding to a distinct color conversion implementation.

17. The system, as in claim 14, generating a set of calibration curves from said halftone classifications.

18. The system, as in claim 14, obtaining at least one mapping between colors observed and an output rendering of CMYK separations.

19. The system, as in claim 16, wherein the pixel-value to color rendering value mappings are tone reproduction curves, each tone reproduction curve corresponding to a distinct color conversion implementation.

20. The system, as in claim 16, wherein the pixel-value to color rendering value mappings are specialized mathematical functions, each specialized mathematical function corresponding to a distinct color conversion implementation.

21. The system, as in claim 14, wherein the pixel-value to color rendering value mappings are matrices, each matrix corresponding to a distinct color conversion implementation.

22. The system, as in claim 14, wherein the pixel-value to color rendering value mappings are sigmoid functions, each sigmoid function corresponding to a distinct color conversion implementation.

23. The system, as in claim 14, further comprising using look-up tables and/or tone reproduction curves to determine a location of at least one boundary region in colorspace.

24. The system, as in claim 23, wherein an optimum boundary region in colorspace is a function of halftone frequency response.

25. The system, as in claim 23, wherein an initial estimate of any one colorspace boundary is preferred until a better estimate of that boundary location is known.

26. A method for obtaining a preferred color conversion implementation for rendering of an image containing halftoned data, comprising:

providing a plurality of pixel-value to color rendering value mappings, each pixel-value to color rendering value mapping corresponding to a distinct color conversion implementation for converting a pixel-value of the image to a pixel-value in a colorspace used for rendering the image, each color conversion implementation corresponding to a distinct halftone type;

processing, using a processor, the image and generating therefrom a plurality of tags denoting the frequencies of halftoned data of the image containing halftoned data;

identifying different classes of halftones from the tags to obtain a plurality of screen classifications;

selecting one of the screen classifications;

selecting one of the plurality of pixel-value to color rendering value mappings, based upon the selected screen classification, to obtain the preferred color conversion implementation for rendering the image containing halftoned data; and

rendering, using a printer, the image containing halftoned data using the selected pixel-value to color rendering value mapping.

27. The method for obtaining a preferred color conversion implementation, as in claim 26, wherein the pixel-value to color rendering value mappings are look-up tables, each look-up table corresponding to a distinct color conversion implementation.

28. The method for obtaining a preferred color conversion implementation, as in claim 26, wherein the pixel-value to color rendering value mappings are tone reproduction curves, each tone reproduction curve corresponding to a distinct color conversion implementation.

29. The method for obtaining a preferred color conversion implementation, as in claim 26, wherein the pixel-value to color rendering value mappings are multi-dimensional.

30. The method for obtaining a preferred color conversion implementation, as in claim 26, wherein the pixel-value to color rendering value mappings are specialized mathematical functions, each specialized mathematical function corresponding to a distinct color conversion implementation.

31. The method, as in claim 26, wherein the pixel-value to color rendering value mappings are matrices, each matrix corresponding to a color.

32. The method, as in claim 26, wherein the pixel-value to color rendering value mappings are sigmoid functions, each sigmoid function corresponding to a distinct color conversion implementation.

33. A system for obtaining a preferred color conversion implementation for rendering of an image containing halftoned data, comprising:

a memory including a plurality of pixel-value to color rendering value mappings, each pixel-value to color rendering value mapping corresponding to a distinct color conversion implementation for converting a pixel-value of the image to a pixel-value in a colorspace used for rendering the image, each color conversion implementation corresponding to a distinct halftone type; and

a processing unit;

said processing unit generating a plurality of tags denoting the frequencies of halftoned data of the image containing halftoned data;

said processing unit identifying different classes of halftones from said tags to obtain a plurality of screen classifications;

said processing unit selecting one of the screen classifications; and

said processing unit selecting, from said memory, one of the plurality of pixel-value to color rendering value mappings, based upon the selected screen classification, to obtain the preferred color conversion implementation for rendering the image containing halftoned data.

34. The system for obtaining a preferred color conversion implementation, as in claim 33, wherein the pixel-value to color rendering value mappings are look-up tables, each look-up table corresponding to a distinct color conversion implementation.

35. The system for obtaining a preferred color conversion implementation, as in claim 33, wherein the pixel-value to color rendering value mappings are tone reproduction curves, each tone reproduction curve corresponding to a distinct color conversion implementation.

36. The system for obtaining a preferred color conversion implementation, as in claim 33, wherein the pixel-value to color rendering value mappings are multi-dimensional.

37. The system for obtaining a preferred color conversion implementation, as in claim 33, wherein the pixel-value to color rendering value mappings are specialized mathematical functions, each specialized mathematical function corresponding to a distinct color conversion implementation.

38. The system, as in claim 33, wherein the pixel-value to color rendering value mappings are matrices, each matrix corresponding to a distinct color conversion implementation.

39. The system, as in claim 33, wherein the pixel-value to color rendering value mappings are sigmoid functions, each sigmoid function corresponding to a distinct color conversion implementation.