Systems and methods for compensating for temperature induced spectral emission variations in LED based color parameter measuring devices
View Patent ↗Systems and method provide compensation for temperature induced peak wavelength shift of LEDs is color parameter measuring systems that use a model to reconstruct target color parameter values from the reflectance values measured when the target is illuminated by LEDs. Several models may be constructed, with each model being trained at a unique temperature, resulting in a set of models that span the temperature range of interest. In real-time, the LED based color parameter measuring system measures the temperature and interpolates between the models to estimate the appropriate model to use at the temperature of interest. The estimated model is then used to perform the color parameter value estimation.
1. A method of determining color parameter values, comprising:
obtaining a plurality of illuminant sensor outputs, each illuminant sensor output obtained by illuminating a target at an operating temperature;
obtaining a measurement of the operating temperature;
retrieving model information from a reference database based on the operating temperature and the sensor outputs, the reference database containing different model information for different temperatures;
determining a model for the operating temperature and the sensor outputs based on the retrieved model information;
determining color parameter values based on the determined model; and
outputting the determined color parameter values to an output media.
2. The method of claim 1 , wherein the color parameter values are reflectance spectra.
3. The method of claim 1 , wherein the color parameter values are L*,a*,b* values.
4. The method of claim 1 , wherein the color parameter values are R,G,B, values.
5. The method of claim 1 , wherein the color parameter values are X,Y,Z, values.
6. The method of claim 1 , further comprising establishing the reference database, the establishing the reference database comprising:
establishing a plurality of temperatures, and for each temperature:
constructing a spectral reconstruction matrix including a plurality of training samples, each training sample correlating a reference spectrum with a corresponding plurality of normalized illuminant sensor outputs for reference colors of the temperature; and
storing the spectral reconstruction matrices in the reference database.
7. The method of claim 1 , wherein
the model information includes matrices for converting illuminant sensor outputs to color parameters, each matrix being trained for a respective temperature,
the retrieved model information includes two matrices corresponding to two temperatures that bracket the operating temperature,
determining the model comprises determining a color parameter vector as:
E
_
^
(
T
)
=
(
1
-
T
-
T
i
T
i
+
1
-
T
i
)
f
(
V
_
;
{
E
_
n
}
,
{
V
_
n
}
,
T
i
)
+
(
T
-
T
i
T
i
+
1
-
T
i
)
f
(
V
_
;
{
E
_
n
}
,
{
V
_
n
}
,
T
i
+
1
)
T
i
≤
T
≤
T
i
+
1
where T is the operating temperature, T i and T i+1 are the two temperatures that bracket the operating temperature, ƒ( V ;{ E n },{ V n },T i ) and ƒ( V ;{ E n },{ V n },T i+1 ) are the retrieved two matrices for temperatures T 1 and T i+1 , respectively, and:
Ê =Vector of Color Parameters
V =Vector of Sensor Measurements
{ E n }=Set of Color Parameter Values used in Device Characterization
{ V n }=Set of Sensor Measurements used in Device Characterization
n =1,2, . . . , N
N=Number of Data Pairs used in Device Characterization.
8. The method of claim 1 , wherein the model information includes matrices for converting illuminant sensor outputs to spectra, each matrix being trained for a respective temperature.
9. The method of claim 8 , wherein
the retrieved model information includes two matrices corresponding to two temperatures that bracket the operating temperature, and
determining the model comprises interpolating the retrieved two matrices to obtain a matrix for the operating temperature.
10. The method of claim 9 , wherein the matrix for the operating temperature is calculated as:
A
*
(
T
)
=
(
1
-
T
-
T
i
T
i
+
1
-
T
i
)
A
*
(
T
i
)
+
(
T
-
T
i
T
i
+
1
-
T
i
)
A
*
(
T
i
+
1
)
,
where T is the operating temperature, A*(T i ) and A*(T i+1 ) are the retrieved two matrices for temperatures T i and T i+1 , respectively, and:
T i ≦T≦T i+1 .
11. The method of claim 8 , wherein
the retrieved model information includes a matrix corresponding to a temperature that is closest to the operating temperature, and
determining the model comprises calculating a matrix for the operating temperature using an extrapolation of the retrieved matrix.
12. The method of claim 11 , wherein the matrix for the operating temperature is calculated as:
A
*
(
T
)
=
A
*
(
T
0
)
+
(
T
-
T
0
T
1
-
T
0
)
(
A
*
(
T
1
)
-
A
*
(
T
0
)
)
=
A
*
(
T
0
)
+
(
∂
A
*
∂
T
)
(
T
-
T
0
)
where T is the operating temperature, A*(T 0 ) is the retrieved matrix for temperature grids T 0 , and
(
∂
A
*
∂
T
)
=
Δ
1
T
1
-
T
0
(
A
*
(
T
1
)
-
A
*
(
T
0
)
)
is a pre-calculated value determined based on a known parameter A*(T 1 ).
13. The method of claim 1 , the reference database containing clusters, each cluster containing a plurality of colors, the method further comprising:
determining an appropriate cluster based on the normalized value,
wherein:
retrieving the model information comprises retrieving model information based on the determined cluster and the operating temperature; and
determining the model comprises determining a model for the determined cluster and the operating temperature.
14. The method of claim 13 , wherein determining the cluster comprises determining a Euclidean distance.
15. A computer-readable medium having computer-executable instructions for performing the method recited in claim 1 .
16. A color parameter value determination system, comprising:
a plurality of illuminants;
at least one photodetector that detects light originating from the plurality of illuminants and reflected by a target; and
a controller that:
processes the plurality of illuminant sensor outputs obtained from at least one photodetector at an operating temperature, each illuminant sensor output indicating a color value obtained from a target;
obtains a measurement of the operating temperature;
retrieves model information from the reference database based on the operating temperature and the sensor outputs, the reference database containing different model information for different temperatures;
determines a model for the operating temperature and the sensor outputs based on the retrieved model information;
determines a color parameter value based on the determined model; and
outputs the determined color parameter value to an output media.
17. The system according to claim 16 , wherein the color parameter values are reflectance spectra.
18. The system according to claim 16 , wherein the color parameter values are L*,a*,b* values.
19. The system according to claim 16 , wherein the color parameter values are R,G,B, values.
20. The system according to claim 16 , wherein the color parameter values are X,Y,Z, values.
21. The system according to claim 16 , further comprising:
a memory that pre-stores the reference database,
wherein the controller establishes a plurality of temperatures, and, for each temperature:
constructs a spectral reconstruction matrix including a plurality of training samples, each training sample correlating a reference spectrum with a corresponding plurality of normalized illuminant sensor outputs for reference colors of the temperature; and
stores the spectral reconstruction matrices in the reference database.
22. The system according to claim 16 , wherein
the model information includes matrices for converting illuminant sensor outputs to color parameters, each matrix being trained for a respective temperature,
the retrieved model information includes two matrices corresponding to two temperatures that bracket the operating temperature,
determining the model comprises determining a color parameter vector as:
E
_
^
(
T
)
=
(
1
-
T
-
T
i
T
i
+
1
-
T
i
)
f
(
V
_
;
{
E
_
n
}
,
{
V
_
n
}
,
T
i
)
+
(
T
-
T
i
T
i
+
1
-
T
i
)
f
(
V
_
;
{
E
_
n
}
,
{
V
_
n
}
,
T
i
+
1
)
T
i
≤
T
≤
T
i
+
1
where T is the operating temperature, T i and T i+1 are the two temperatures that bracket the operating temperature, ƒ( V ;{ E n }, { V n },T i ) and ƒ( V ;{ E n }, { V n },T i+1 ) are the retrieved two matrices for temperatures T i and T i+1 , respectively, and:
Ê =Vector of Color Parameters
V =Vector of Sensor Measurements
{ E n }=Set of Color Parameter Values used in Device Characterization
{ V n }=Set of Sensor Measurements used in Device Characterization
n =1,2, . . . , N
N =Number of Data Pairs used in Device Characterization.
23. The system of claim 16 , wherein the model information includes matrices for converting illuminant sensor outputs to spectra, each matrix being trained for a temperature.
24. The system of claim 23 , wherein
the retrieved model information includes two matrices corresponding to two temperature grid points that bracket the operating temperature, and
the controller interpolates the retrieved two matrices to obtain a matrix for the operating temperature.
25. The system of claim 24 , wherein the matrix for the operating temperature is calculated as:
A
*
(
T
)
=
(
1
-
T
-
T
i
T
i
+
1
-
T
i
)
A
*
(
T
i
)
+
(
T
-
T
i
T
i
+
1
-
T
i
)
A
*
(
T
i
+
1
)
,
where T is the operating temperature, A*(T i ) and A*(T i+1 ) are the retrieved two matrices for temperature grid points T 1 and T i+1 , respectively, and:
T i ≦T≦T i+1 .
26. The system of claim 23 , wherein
the retrieved model information includes a matrix corresponding to a temperature grid point that is closest to the operating temperature, and
the controller calculates a matrix for the operating temperature using an extrapolation of the retrieved matrix.
27. The system of claim 26 , wherein the matrix for the operating temperature is calculated as:
A
*
(
T
)
=
A
*
(
T
0
)
+
(
T
-
T
0
T
1
-
T
0
)
(
A
*
(
T
1
)
-
A
*
(
T
0
)
)
=
A
*
(
T
0
)
+
(
∂
A
*
∂
T
)
(
T
-
T
0
)
where T is the operating temperature, A*(T 0 ) is the retrieved matrix for temperature grid points T 0 , and
(
∂
A
*
∂
T
)
=
Δ
1
T
1
-
T
0
(
A
*
(
T
1
)
-
A
*
(
T
0
)
)
is a pre-calculated value determined based on a known parameter A*(T1) and T1.
28. The system of claim 16 , the reference database containing clusters, each cluster containing a plurality of colors, wherein the controller:
determines an appropriate cluster based on the normalized value;
retrieves the model information based on the determined cluster and the operating temperature; and
determines the model for both the determined cluster and the operating temperature.
29. The system of claim 28 , wherein the controller determines a Euclidean distance.
30. A coloring system incorporating the system of claim 16 .
31. The coloring system of claim 30 , wherein the coloring system is one of a digital photocopier and a color printer.
32. The coloring system of claim 30 , wherein the coloring system is a xerographic color printer.
33. The coloring system of claim 30 , wherein the coloring system is an ink-jet printer or solid ink-jet printer.
34. A color detection system incorporating the system of claim 16 .