Appearance-based technique for rendering colors on an output device

• US 5,650,942 A
• Filed: 02/05/1996
• Issued: 07/22/1997
• Est. Priority Date: 02/02/1996
• Status: Expired due to Term

• Current Assignee: X-Rite Incorporated (Danaher Corporation)
• Original Assignee: Light Source Computer Images, Inc. (Danaher Corporation)
• Inventors: Granger, Edward M.
• Primary Examiner(s): Voeltz, Emanuel T.
• Assistant Examiner(s): DAM, TUAN QUANG
• Application Number:
  US08/595,404
• Time in Patent Office: 533 Days
• Field of Search: 364/526, 364/927.3, 364/930.4, 358/501, 358/502, 358/518, 358/520, 358/521, 358/523, 358/524-525, 358/529, 395/101, 395/109, 395/118, 395/127, 395/129, 395/131, 395/132, 345/112, 345/150, 345/153, 345/154-155, 345/185, 345/189, 345/199, 345/202
• US Class Current: 358/500
• CPC Class Codes: H04N 1/60 Colour correction or control H04N1/54 takes precedence
• CPC Class Codes: H04N 1/6058 Reduction of colour to a range of reproducible colours, e.g. to ink- reproducible colour gamut

Abstract

A target output device such as a printer is characterized in a manner that primary colorant amounts can be determined on the basis of a pair of chromaticity coordinates in a chromaticity space having a pair of chromaticity coordinates that span hue and chroma (saturation) and a luminance correlate (monotonic function of luminance). Lightness (or darkness) of the colors is handled largely independently. This makes it possible to store the colorant information in a two-dimensional table indexed by chromaticity coordinates. The information in the table is for colors that are characterized by the highest luminance that the target printer is capable of rendering for a given chromaticity (hue and chroma).

Claims

1. A computer-implemented method of determining colorant values for rendering a color on a target printer wherein

the color is specified by a set of values that include a lightness value LL_col that correlates with luminance and a pair of chromaticity values c1_col and c2_col in a two-dimensional space that spans hue and chroma, and

the target printer responds to commands specifying amounts of a set of colorants including at least two primary colorants, the method comprising the steps of:

determining, as a first function of c1_col and c2_col, (a) a first set of prescribed colorant values for printing a color characterized by the c1_col and c2_col pair at the maximum lightness that the target printer can print for c1_col and c2_col, and (b) a representation of the maximum lightness, referred to as LL_top, for c1_col and c2_col ;

determining a darkness value D_col of the color on the basis of LL_col and the representation of LL_top for c1_col and c2_col ;

determining, as a second function of the darkness value, D_col, a second set of colorant values;

for at least some values of D_col, adjusting the first set of prescribed colorant values downwardly; and

determining a third set of colorant values by combining the first and second sets of colorant values.

2. The method of claim 1 wherein:

the first function is implemented by providing a two-dimensional table, referred to as the umbrella table, characterizing the target printer, which table includes, for each particular one of a plurality of c1-c2 pairs, an entry including (a) a set of colorant values for printing a color characterized by the particular c1-c2 pair at the maximum lightness that the target printer can print for the particular c1-c2 pair, and (b) a representation of the maximum lightness, referred to as LL_top, for the particular c1-c2 pair; and

said step of determining as a first function includes using the umbrella table to find (a) said first set of prescribed colorant values for printing a color characterized by c1_col and c2_col at the maximum lightness, and (b) said representation of LL_top for c1_col and c2_col.

3. The method of claim 1 wherein:

the second function is implemented by providing a table, referred to as the darkness table, which table includes for each particular one of a plurality of specified darkness values, an entry including a set of colorant values for printing a neutral component having an actual darkness that bears a predetermined relationship to the particular specified darkness value.

4. The method of claim 3 wherein said predetermined relationship is that the actual darkness is substantially equal to the specified darkness for darkness values up to a threshold darkness, and smoothly decreases from the specified darkness to a value equal to the maximum darkness that the target printer can print when the specified darkness is 100%.

5. A computer-implemented method of determining colorant values for rendering a color on a target printer wherein

the color is specified by a set of values that include a lightness value LL_col that correlates with luminance and a pair of chromaticity values c1_col and c2_col in a two-dimensional space that spans hue and chroma, and

the target printer responds to commands specifying amounts of a set of colorants including at least two primary colorants,

the method comprising the steps of:

providing a two-dimensional table, referred to as the umbrella table, characterizing the target printer, which table includes, for each particular one of a plurality of c1-c2 pairs, an entry including (a) a set of colorant values for printing a color characterized by the particular c1-c2 pair at the maximum lightness that the target printer can print for the particular c1-c2 pair, and (b) a representation of the maximum lightness, referred to as LL_top, for the particular c1-c2 pair;

determining, on the basis of the umbrella table, (a) a first set of prescribed colorant values for printing a color characterized by c1_col and c2_col at the maximum lightness, and (b) a representation of LL_top for c1_col and c2_col ;

providing a table, referred to as the darkness table, which table includes for each particular one of a plurality of darkness values, an entry including a set of colorant values for printing a neutral component having a darkness that bears a predetermined relationship to the particular darkness value;

determining a darkness value D_col of the color on the basis of LL_col and the representation of LL_top for c1_col and c2_col ;

for at least some values of D_col, adjusting the first set of prescribed colorant values downwardly;

determining a second set of colorant values from the darkness value D_col and the darkness tables; and

determining a third set of colorant values by combining the first and second sets of colorant values.

6. The method of claim 5 wherein said step of adjusting the first set of colorant values comprises the steps of:

determining as a function of the darkness value D_col a chroma reduction factor;

determining a pair of values c1_col, and c2_col, that define a color having the same hue as defined by c1_col and c2_col but a chroma reduced by the chroma reduction factor;

determining, on the basis of the umbrella table, a revised set of prescribed colorant values for printing a color characterized by c1_col, and c2_col, at the maximum lightness, the revised set of colorant values providing the downwardly adjusted prescribed colorant values.

7. The method of claim 5 wherein:

the umbrella table includes entries for substantially all possible c1-c2 pairs; and

said step of determining on the basis of the umbrella table includes extracting the first set of prescribed colorant values and the representation of LL_top for c1_col and c2_col directly from an entry in the umbrella table.

8. The method of claim 5 wherein:

the umbrella table includes entries for substantially fewer than all possible c1-c2 pairs; and

said step of determining on the basis of the umbrella table includes extracting a number of entries from the umbrella table, and deriving the first set of prescribed colorant values and the representation of LL_top for c1_col and c2_col by interpolation of values from the entries, so extracted.

9. The method of claim 5 wherein said step of providing the umbrella table comprises the steps of:

printing a set of color samples with the target printer using known colorant values, the color of each sample being substantially devoid of a neutral component so as to provide the lightest color that the target printer is capable of printing at the given hue and chroma;

measuring each color sample to provide values of c1, c2, and lightness LL for that color sample; and

associating with the measured c1 and c2 values for at least some of the color samples, the colorant values that produced that color sample, and the measured lightness value for that color sample.

10. The method of claim 5 wherein:

R, G, and B are a set of tristimulus values related to CIE XYZ values by a predetermined linear transformation characterized by a non-singular 3×

3 transformation matrix;

the pair of chromaticity values, denoted t and d, are related to the RGB values according to the equations

space="preserve" listing-type="equation">t=2(R-G)/(2R+3G+2B) and

space="preserve" listing-type="equation">d=(R+G-2B)/(2R+3G+2B); and

the lightness value, designated V, is given by the equation

space="preserve" listing-type="equation">V=(2R+3G+2B)/2.

11. The method of claim 10 wherein:

the pair of chromaticity values are derived from t and d according to the equations

space="preserve" listing-type="equation">r=(t.sup.2 +d.sup.2).sup.1/2 ; and

space="preserve" listing-type="equation">θ

=tan.sup.-1 (t/d).

12. The method of claim 5 wherein:

the set of colorants includes exactly three primary colorants; and

the primary colorants are C, M, and Y ink densities.

13. The method of claim 5 wherein the set of color samples includes only colors rendered by single primary colorants and pairwise combinations of adjacent primary colorants.

14. The method of claim 5 wherein the set of color samples for a given adjacent pair of primary colorants includes:

a sample with a maximum density of one primary colorant with none of the other primary colorant;

a sample with a maximum density of the other primary colorant with none of the one primary colorant; and

a sample with a maximum density of both primary colorants.

15. A computer implemented method of determining colorant values for rendering a color specified by a set of tristimulus values to an output device that responds to commands specifying at least two primary colorants and a form of neutral colorant, the method comprising the steps of:

(a) mapping the tristimulus values to a set of values that include a lightness value LL that correlates with luminarice and a pair of chromaticity values c1 and c2 in a two-dimensional space that spans hue and chroma;

(b) deriving a set of parameters as a first function of the chromaticity values only, the set of parameters including primary colorant values; and

(c) deriving the desired values of the neutral colorant as a second function of the set of parameters and the lightness value.

16. The method of claim 15 wherein:

the set of parameters includes a set P of primary colorant values and a value LL_top representing the lightness value of the color represented by P.

17. The method of claim 16 wherein:

the second function depends only on the value of LL_top and LL_col.

18. The method of claim 16, and further comprising the steps of:

further comparing LL_top and LL_col ;

if LL_col is greater than LL_top, modifying c1 and c2 to maintain hue and reduce chroma so that when the first function is applied to the modified c1 and c2, the resultant modified value of LL_top is substantially equal to LL_col.

19. The method of claim 16 wherein:

the set P of primary colorant values includes no more than two non-zero values.

20. The method of claim 15 wherein:

the first function uses a table indexed by the pair of chromaticity values.

21. The method of claim 20 wherein:

c1 and c2 have n1 and n2 bits of precision, respectively;

the table contains 2.sup.(n1+n2) entries corresponding to all possible combinations of values for c1 and c2; and

the first function is implemented by a direct lookup in the table.

22. The method of claim 20 wherein:

c1 and c2 have n1 and n2 bits of precision, respectively;

the table contains substantially fewer than 2.sup.(n1+n2) entries; and

the first function is implemented by interpolating values extracted from the table.

23. The method of claim 15 wherein:

the tristimulus values are related to CIE XYZ values by a predetermined linear transformation characterized by a non-singular 3×

3 transformation matrix.

24. The method of claim 15 wherein:

the set of tristimulus values are defined relative to a set of three primaries, designated the meta RGB primaries, that lie on or outside the normal CIE (x, y) chromaticity boundary.

25. The method of claim 15 wherein:

the set of tristimulus values are a set of RGB values related to CIE XYZ values by a predetermined linear transformation characterized by a non-singular 3×

3 transformation matrix;

the pair of chromaticity values, denoted t and d, are related to the RGB values according to the equations

space="preserve" listing-type="equation">t=2(R-G)/(2R+3G+2B) and

space="preserve" listing-type="equation">d=(R+G-2B)/(2R+3G+2B); and

the lightness value, designated V, is given by the equation

space="preserve" listing-type="equation">V=(2R+3G+2B)/2.

26. The method of claim 15 wherein:

the pair of chromaticity values are derived from t and d according to the equations

space="preserve" listing-type="equation">r=(t.sup.2 +d.sup.2).sup.1/2 ; and

space="preserve" listing-type="equation">θ

=tan.sup.-1 (t/d).

27. A memory for storing data for access by an application program being executed on a data processing system, the program operating to determine colorant values for rendering, on a target output device, colors specified by a pair of chromaticity values c1_col and c2_col in a two-dimensional space that spans hue and chroma, the memory comprising:

a data structure stored in said memory, said data structure including, for each particular one of a plurality of c1-c2 pairs, an entry including

(a) a representation of a set of colorant values for printing a color characterized by the particular c1-c2 pair at the maximum lightness that the target printer can print for the particular c1-c2 pair, and

(b) a representation of the maximum lightness, referred to as LL_top, for the particular c1-c2 pair.

28. The memory of claim 27 wherein said data structure is organized as at least one table indexed by the values of the c1-c2 pair.

29. The memory of claim 27 wherein each entry includes:

colorant values for a pair of adjacent primary colorants of a plurality of at least three primary colorants; and

only two colorant values; and

an indicator of which pair of primary colorants.

30. A memory for storing data for access by an application program being executed on a data processing system, the program operating to determine colorant values for darkening a rendered color on a target output device, the memory comprising:

a data structure stored in said memory, said data structure including, for each particular one of a plurality of darkness values, an entry including

(a) a representation of a set of colorant values for printing a neutral component having a darkness that bears a predetermined relationship to the particular darkness value, and

(b) a representation of a chroma reduction factor for the color to be rendered.

31. A method of characterizing a target printer, the target printer being responsive to commands specifying values for at least one of a plurality of primary colorants, the method comprising the steps of:

printing a set of color samples with the target printer using known colorant values substantially devoid of a neutral component so as to provide the lightest color that the target printer is capable of printing at the given hue and chroma;

measuring each color sample to provide a measurement of the form c1, c2, LL where LL is a lightness value that correlates with luminance and c1 and c2 are a pair of chromaticity values in a two-dimensional space that spans hue and chroma; and

associating with the measured c1 and c2 values for at least some of the color samples, the colorant values that produced that color sample, and the measured lightness value for that color sample.

32. The method of claim 31, and further comprising the steps of:

constructing a table containing, for each of a set of evenly spaced values for c1 and c2, interpolated colorant commands and an interpolated lightness value derived from said associating step.

33. The method of claim 31 wherein:

the plurality of primary colorants includes exactly three primary colorants; and

the primary colorants are C, M, and Y ink densities.

34. The method of claim 31 wherein the set of color samples includes only colors rendered by single primary colorants and pairwise combinations of adjacent primary colorants.

35. The method of claim 31 wherein the set of color samples for a given pair of primary colorants includes:

a sample with a maximum density of one primary colorant with none of the other primary colorant;

a sample with a maximum density of the other primary colorant with none of the one primary colorant; and

a sample with a maximum density of both primary colorants.

36. The method of claim 31 wherein said measuring step includes the substeps of steps of:

defining an intermediate color space, referred to as the meta RGB space, having three meta RGB coordinates related to CIE XYZ coordinates by a linear transformation characterized by a non-singular 3×

3 transformation matrix;

providing a color measurement device capable of providing CIE XYZ values for a given measurement;

measuring each color sample with the color measurement device;

converting the measured CIE XYZ values for each color sample to meta RGB values; and

converting the meta RGB values to c1, c2, LL values.

37. The method of claim 36 wherein said step of defining the meta RGB space includes the substeps of:

determining X, Y, and Z values for a set of three initial RGB primaries and a reference white point;

transforming the three initial RGB primaries to reference RGB primaries by proportionately increasing the distances of the initial RGB primaries from the reference white point so as to span all colors to be represented in meta space; and

determining a transformation from X, Y, and Z values to meta RGB coordinates that maps the three reference RGB primaries to respective points along the meta RGB coordinate axes with maximal coordinate values and the reference white point to a point in meta space having all three coordinates with maximal coordinate values.

38. The method of claim 36 wherein:

the pair of chromaticity values, denoted t and d, are related to the meta RGB values according to the equations

space="preserve" listing-type="equation">t=2(R-G)/(2R+3G+2B) and

space="preserve" listing-type="equation">d=(R+G-2B)/(2R+3G+2B); and

the lightness value, designated V, is given by the equation

space="preserve" listing-type="equation">V=(2R+3G+2B)/2.

39. The method of claim 38 wherein:

the pair of chromaticity values are derived from t and d according to the equations

space="preserve" listing-type="equation">r=(t.sup.2 +d.sup.2).sup.1/2 ; and

space="preserve" listing-type="equation">θ

=tan.sup.-1 (t/d).

40. A method of printing a color image on a reproducing printer having a first set of colorants with a first set of maximum chroma values so as to simulate the appearance of the same image on an original printer having a second set of colorants with a second set of maximum chroma values greater than the first set of maximum chroma values, the method comprising:

defining a function of specified chroma value that is substantially linear with unit slope up to a threshold chroma value, and falls off gradually to a value less than 100% for a specified chroma value of 100%, the value representing the ratio of one of the first maximum chroma values to a corresponding one of the second maximum chroma values;

in response to specified hue and chroma values to be rendered by the reproducing printer, applying the function to the specified chroma value;

determining colorant amounts for a color having the specified hue and the specified chroma, as possibly reduced by the function; and

rendering a pixel in the image using the colorant amounts so determined.

41. A method of printing a color image on a reproducing printer having a first set of colorants so as to simulate the appearance of the same image on an original printer having a second set of colorants, the second set of colorants allowing a greater darkness value than allowed by the first set of colorants, the method comprising:

defining a function of specified darkness value that is substantially linear with unit slope up to a threshold darkness value, and falls off gradually to a value less than 100% for a specified darkness value of 100%, the value representing the ratio of first maximum darkness value to a the second maximum darkness value;

in response to specified values to be rendered by the reproducing printer, applying the function to the specified darkness value;

determining colorant amounts for a neutral color having the specified darkness, as possibly reduced by the function; and

rendering a pixel in the image using the colorant amounts so determined for a neutral component.