Color processing

• US 6,608,925 B1
• Filed: 03/01/1999
• Issued: 08/19/2003
• Est. Priority Date: 03/01/1999
• Status: Expired due to Fees

• Current Assignee: Eastman Kodak Company
• Original Assignee: Kodak Polychrome Graphics Company Limited (Eastman Kodak Company)
• Inventors: Fischer, Timothy A., Rozzi, William A., Edge, Christopher J.
• Primary Examiner(s): Tran, Phuoc
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US09/259,579
• Time in Patent Office: 1,632 Days
• Field of Search: 382/162, 382/167, 382/274, 358/518, 358/520, 358/521, 358/519, 358/522, 358/516, 345/1.9, 345/589-591, 345/597, 345/600-604, 348/603, 348/655, 348/223.1, 348/708
• US Class Current: 382/162
• CPC Class Codes: H04N 1/6027 Correction or control of colour gradation or colour contrast H04N1/6058 takes precedence
• CPC Class Codes: H04N 1/6052 Matching two or more picture signal generators or two or more picture reproducers

Abstract

A method of characterizing a color imaging system is provided. The method comprises obtaining first data indicative of output of the color imaging system. The first data is processed, to yield second data, according to a color appearance model that varies in accordance with neutrality of colors indicated by the first data. Other methods are provided as well as systems and computer program products for characterizing color imaging systems and devices and for producing colors.

Claims

1. A method of characterizing a color imaging system, the method comprising:

obtaining first data indicative of output of the color imaging system;

processing the first data, to yield second data, according to a color appearance model that varies in accordance with neutrality of colors indicated by the first data, wherein the color appearance model varies according to a white reference vector that is a weighted combination of a local white point of the color imaging system and a common white point, the white reference vector being weighted more to the local white point the more a color indicated by the first data is neutral and being weighted more to the common white point the more the indicated color is saturated.

2. The method of claim 1 wherein the color imaging system is an emissive system and processing the first data includes using a media white point as the local white point to implement absolute colorimetry.

3. The method of claim 1 wherein the color appearance model varies as a function of intensity of the color indicated by the first data.

4. The method of claim 3 wherein the color appearance model includes a luminance descriptor, and a pair of color descriptors that quantify relative amounts of red, green, yellow, and blue in a color indicated by the second data;

wherein the luminance descriptor varies as a function of Y, Y being one of tristimulus values X, Y, and Z of the color indicated by the first data; and

wherein the pair of color descriptors vary as functions of the neutrality of the color indicated by the first data.

5. The method of claim 4 wherein the luminance descriptor varies as a function of a Y-reference that is a weighted combination of a local white point Y-value and a common white point Y-value, the Y-reference being weighted more toward the local white point Y-value the closer the Y-value of the color indicated by the first data is to the local white point Y-value and being weighted more toward the common white point Y-value the more the Y-value of the color indicated by the first data and the local white point Y-value differ.

6. The method of claim 5 wherein the second data include values for L⁺, a⁺, and b⁺;

and

wherein $L^{+}=116\times <br><br>f<br><br>\left(Y/Y_n^{\mathrm{″<br><br>}}\right)-16<br><br><br><br>Y_n^{\mathrm{″<br><br>}}=Y_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)\right)+Y_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)<br><br><br><br>\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)=1.0-\left(Y/Y_{\mathrm{LW}}\right)<br><br><br><br>a^{+}=500<br><br>\left(f<br><br>\left(X/X_n^{\mathrm{\prime <br><br>}}\right)-f<br><br>\left(Y/Y_n^{\mathrm{\prime <br><br>}}\right)\right)<br><br><br><br>b^{+}=200<br><br>\left(f<br><br>\left(Y/Y_n^{\mathrm{\prime <br><br>}}\right)-f<br><br>\left(Z/Z_n^{\mathrm{\prime <br><br>}}\right)\right)<br><br><br><br>f<br><br>\left(\mathrm{\omega <br><br>}\right)={\left(\mathrm{\omega <br><br>}\right)}^{1/3}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\omega <br><br>}><br><br>0.008856<br><br><br><br>f<br><br>\left(\mathrm{\omega <br><br>}\right)=7.787<br><br>\left(\mathrm{\omega <br><br>}+16/116<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\omega <br><br>}\le <br><br>0.008856<br><br><br><br>X_n^{\mathrm{\prime <br><br>}}=X_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+X_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)<br><br><br><br>Y_n^{\mathrm{\prime <br><br>}}=Y_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+Y_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right))$$Z_n^{\mathrm{\prime <br><br>}}=Z_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+Z_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right))$$C=\left(X,Y,Z\right)$$C_{\mathrm{LW}}=\left(X_{\mathrm{LW}},Y_{\mathrm{LW}},Z_{\mathrm{LW}}\right)$$\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)={\left({\mathrm{devX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}/\max <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Dev}\right)}^{\mathrm{\gamma <br><br>}}$$\max <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Dev}=\mathrm{sqrt}<br><br>\left(6.0/9.0\right)*\max <br><br>\left(X^{\mathrm{\prime <br><br>}},Y^{\mathrm{\prime <br><br>}},Z^{\mathrm{\prime <br><br>}}\right)$${\mathrm{devX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}=\mathrm{sqrt}<br><br>\left({\left(X^{\mathrm{\prime <br><br>}}-{\mathrm{avgX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2+{\left(Y^{\mathrm{\prime <br><br>}}-{\mathrm{avgX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2+{\left(Z^{\mathrm{\prime <br><br>}}-{\mathrm{avgX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2\right)$${\mathrm{avgX}}^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}=\left(X^{\mathrm{\prime <br><br>}}+Y^{\mathrm{\prime <br><br>}}+Z^{\mathrm{\prime <br><br>}}\right)/3.0$$X^{\mathrm{\prime <br><br>}}=X/X_{\mathrm{LW}}$$Y^{\mathrm{\prime <br><br>}}=Y/Y_{\mathrm{LW}}$$Z^{\mathrm{\prime <br><br>}}=Z/Z_{\mathrm{LW}}$

where C_LW is a local white vector representing a local white point of the system, C_CW=(X_CW,Y_CW,Z_CW) is a common white vector for a common white point of the system, and γ

is a variable for scaling the local white vector C_LW relative to the common white vector C_CW.

7. The method of claim 4 wherein the second data include values for L⁺, a⁺, and b⁺;

wherein L⁺ is closer to a relative calorimetric value of L* than an absolute calorimetric value of L* the closer the value of Y is to a local white point value Y_LW; and

wherein a⁺ and b⁺ are closer to relative calorimetric values of a* and b*, respectively, than to absolute calorimetric values of a* and b*, respectively, the closer the indicated color is to neutral.

8. The method of claim 7 wherein $L^{+}=(1.0-\mathrm{sat}$

 



_

 



L*)*L*rel+sat

 



_

 



L**L*abs;

$a^{+}=\left(1.0-\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{*}<br><br>b^{*}\right)*{a^{*}}_{\mathrm{rel}}+\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{*}<br><br>b^{*}*{a^{*}}_{\mathrm{abs}};<br><br>$$b^{+}=\left(1.0-\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{*}<br><br>b^{*}\right)*{b^{*}}_{\mathrm{rel}}+\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{*}<br><br>b^{*}*{b^{*}}_{\mathrm{abs}};<br><br>$$\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>L^{*}=1.0-\left(Y/Y_{\mathrm{LW}}\right);<br><br>$$\mathrm{sat}<br><br>\text{\hspace{1em}<br><br>}<br><br>\_<br><br>\text{\hspace{1em}<br><br>}<br><br>a^{*}<br><br>b^{*}=\left(\mathrm{sqrt}<br><br>\left(a^{*2}+b^{*2}\right)\right)/L^{*};<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{and}$

wherein L*_rel, a*_rel, and b*_rel are values of L*, a*, and b*, respectively, using relative colorimetry, and L*_abs, a*_abs, and b*_abs are values of L*, a*, and b*, respectively, using absolute colorimetry.

9. A computer program product residing on a computer readable medium, for characterizing a color imaging system, comprising instructions for causing a computer to:

obtain first data indicative of output of the color imaging system;

process the first data, to yield second data, according to a color appearance model that varies in accordance with neutrality of a color indicated by the first data, wherein the color appearance model varies according to a white reference vector that is a weighted combination of a local white point of the color imaging system and a common white point the white reference vector being weighted more to the local white point the more a color indicated by the first data is neutral and being weighted more to the common white point the more the indicated color is saturated.

10. The computer program product of claim 9 wherein the color imaging system is an emissive system and the instructions for causing the computer to process the first data cause the computer to use a media white point as the local white point to implement absolute colorimetry.

11. The computer program product of claim 9 wherein the color appearance model varies as a function of intensity of the color indicated by the first data.

12. The computer program product of claim 11 wherein the color appearance model includes a luminance descriptor, and a pair of color descriptors that quantify relative amounts of red, green, yellow, and blue in a color indicated by the second data;

wherein the luminance descriptor varies as a function of Y, Y being one of tristimulus values X, Y, and Z of the color indicated by the first data; and

wherein the pair of color descriptors vary as functions of the neutrality of the color indicated by the first data.

13. The computer program product of claim 12 wherein the luminance descriptor varies as a function of a Y-reference that is a weighted combination of a local white point Y-value and a common white point Y-value, the Y-reference being weighted more toward the local white point Y-value the closer the Y-value of the color indicated by the first data is to the local white point Y-value and being weighted more toward the common white point Y-value the more the Y-value of the color indicated by the first data and the local white point Y-value differ.

14. The computer program product of claim 13 wherein the second data include values for L⁺, a⁺, and b⁺;

and

wherein $L^{+}=116\times <br><br>f<br><br>\left(Y/Y_n^{\mathrm{″<br><br>}}\right)-16<br><br><br><br>Y_n^{\mathrm{″<br><br>}}=Y_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)\right)+Y_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)<br><br><br><br>\mathrm{sat}<br><br>\left(Y,Y_{\mathrm{LW}}\right)=1.0-\left(Y,Y_{\mathrm{LW}}\right)<br><br>\text{\hspace{1em}<br><br>}<br><br><br><br>a^{+}=500<br><br>\left(f<br><br>\left(X/X_n^{\mathrm{\prime <br><br>}}\right)-f<br><br>\left(Y/Y_n^{\mathrm{\prime <br><br>}}\right)\right)<br><br><br><br>b^{+}=200<br><br>\left(f<br><br>\left(Y/Y_n^{\mathrm{\prime <br><br>}}\right)-f<br><br>\left(Z/Z_n^{\mathrm{\prime <br><br>}}\right)\right)<br><br><br><br>f<br><br>\left(\mathrm{\omega <br><br>}\right)={\left(\mathrm{\omega <br><br>}\right)}^{1/3}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\omega <br><br>}><br><br>0.008856<br><br><br><br>f<br><br>\left(\mathrm{\omega <br><br>}\right)=7.787<br><br>\left(\mathrm{\omega <br><br>}\right)+16/116<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\omega <br><br>}\le <br><br>0.008856<br><br><br><br>X_n^{\mathrm{\prime <br><br>}}=X_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+X_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right))<br><br><br><br>Y_n^{\mathrm{\prime <br><br>}}=Y_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+Y_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right))<br><br><br><br>Z_n^{\mathrm{\prime <br><br>}}=Z_{\mathrm{LW}}<br><br>\left(1-\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)\right)+Z_{\mathrm{CW}}*\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right))$$C=\left(X,Y,Z\right)$$C_{\mathrm{LW}}=\left(X_{\mathrm{LW}},Y_{\mathrm{LW}},Z_{\mathrm{LW}}\right)$$\mathrm{sat}<br><br>\left(C,C_{\mathrm{LW}}\right)={\left(\mathrm{dev}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}/\max <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Dev}\right)}^{\mathrm{\gamma <br><br>}}$$\max <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Dev}=\mathrm{sqrt}<br><br>\left(6.0/9.0\right)*\max <br><br>\left(X^{\mathrm{\prime <br><br>}},Y^{\mathrm{\prime <br><br>}},Z^{\mathrm{\prime <br><br>}}\right)$$\mathrm{dev}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}=\mathrm{sqrt}<br><br>\left({\left(X^{\mathrm{\prime <br><br>}}-\mathrm{avg}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2+{\left(Y^{\mathrm{\prime <br><br>}}-\mathrm{avg}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2+{\left(Z^{\mathrm{\prime <br><br>}}-\mathrm{avg}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}\right)}^2\right)$$\mathrm{avg}<br><br>\text{\hspace{1em}<br><br>}<br><br>X^{\mathrm{\prime <br><br>}}<br><br>Y^{\mathrm{\prime <br><br>}}<br><br>Z^{\mathrm{\prime <br><br>}}=\left(X^{\mathrm{\prime <br><br>}}+Y^{\mathrm{\prime <br><br>}}+Z^{\mathrm{\prime <br><br>}}\right)/3.0$$X^{\mathrm{\prime <br><br>}}=X/X_{\mathrm{LW}}$$Y^{\mathrm{\prime <br><br>}}=Y/Y_{\mathrm{LW}}$$Z^{\mathrm{\prime <br><br>}}=Z/Z_{\mathrm{LW}}$

where C_LW is a local white vector representing a local white point of the system, C_CW=(X_CW,Y_CW,Z_CW) is a common white vector for a common white point of the system, and γ

is a variable for scaling the local white vector C_LW relative to the common white vector C_CW.

15. A method of producing a color with an emissive device using absolute colorimetry, the method comprising:

obtaining first data indicative of a first color;

determining second data related to the first data by a color appearance model that uses a white point of the emissive device as a white reference vector; and

actuating the emissive device according to the second data to implement absolute colorimetry to produce a second color to approximate the first color, wherein the white reference vector varies from the white point of the emissive device when the second color is near white to a common white reference, different from the white point of the emissive device, when the second color departs from a near-white, neutral color.

16. A method of characterizing an emissive device for absolute colorimetry, the method comprising:

obtaining first data indicative of output of the emissive device;

converting the first data to second data using a color appearance model that uses a white point of the emissive device as a reference white vector;

providing the second data for use in absolute calorimetric color reproduction, wherein converting the first data to second data further includes using, as the white reference vector, a composite white reference vector that is a weighted combination of the white point of the emissive device and a predetermined white point, the composite white reference vector being closer to the white point of the emissive device the closer a color indicated by the first data is to being neutral.

17. A computer program product residing on a computer readable medium, for characterizing an emissive device for absolute colorimetry, comprising instructions for causing a computer to:

obtain first data indicative of output of the emissive device;

convert the first data to second data using a color space that uses a white point of the emissive device as a reference white vector; and

provide the second data for use in absolute colorimetric color reproduction, wherein the instructions cause the computer to convert the first data to second data using, as the white reference vector, a composite white reference vector that is a weighted combination of the white point of he emissive device and a predetermined white point, the composite white reference vector being closer to the white point of the emissive device the closer a color indicated by the first data is to being neutral.