McolSetMethod

Filters
See also

User Guide:

Advanced color matching settings and concepts Color matching Color processing and analysis examples Color spaces and converting between them Dealing with color Distance between colors Relative color calibration

Reference:

McolAlloc McolControl McolDefine McolInquire McolMatch McolPreprocess McolTransform

Availability

Not available in MIL-Lite

Available in MIL

Available on Windows

Available on Linux

Available on Non-Matrox computer

Available on Matrox 4Sight-X

Available on Matrox 4Sight GP/GPm

Available on Matrox Supersight

Function map

Examples

cielabconversion.cpp
colorbasedproductidentification.cpp
colorrelativecalibration.cpp
foodinspection_cereal.cpp
foodinspectioncereal.cpp
mcol.cpp
mcol.cs
mcol.vb
productidentificationoasis.cpp

Function Map

- ContextId

Table:	For specifying the distance operation (color matching)
Table:	For specifying the mapping operation (relative color calibration)

- DistTypeOrCalIntent

- ConversionModeOrComputeOption

- ControlFlag

Parameter Associations

Table:	For specifying the type of distance calculation (color matching)
Table:	For specifying the color calibration intent (quality) of the mapping operation (relative color calibration)

Compilation information

Synopsis

Set the color matching or relative color calibration strategy.

Syntax

void McolSetMethod(

MIL_ID ContextId,	//in
MIL_INT64 OperationMode,	//in
MIL_INT64 DistTypeOrCalIntent,	//in
MIL_INT64 ConversionModeOrComputeOption,	//in
MIL_INT64 ControlFlag	//in

)

Description

This function sets the strategy with which to perform color matching or relative color calibration, when calling McolMatch() or McolTransform(), respectively. The strategy is based on the operation and either the color distance (for color matching) or the color calibration intent (for relative color calibration). By manipulating these settings, you can affect the resulting color-samples that MIL matches to the target areas, or the color-sample mappings that MIL uses to transform color data with relative color calibration.

If you perform color matching using an RGB source color space, you can use this function to convert your RGB color data to a different color space before the match operation. To do so, explicitly specify a conversion mode (ConversionModeOrComputeOption).

The strategy set with this function is part of the specified context. For this function's settings to take effect, you must preprocess the context (McolPreprocess()).

Note that this function reference has not been updated for a MIL system added during a MIL update. Refer to the MIL system's release note to see if there is complementary information.

Parameters

This function is not supported on the selected boards.

Parameters

ContextId

Specifies the identifier of the color matching or relative color calibration context. The context must have been previously allocated on the required system using McolAlloc() with M_COLOR_MATCHING or M_COLOR_CALIBRATION_RELATIVE.

See the Parameter associations section for possible values that can be specified.

OperationMode

Specifies the operation.

For specifying the operation with which to perform McolMatch(), set this parameter to one of the following values. The ContextId parameter must indicate the identifier of a color matching context.

For specifying the distance operation (color matching) INQ
Value	Description
M_DEFAULT	Same as M_STAT_MIN_DIST.
M_HISTOGRAM_MATCHING	Specifies a minimum distance operation, based on color histograms. When using M_HISTOGRAM_MATCHING, you must set M_DISTANCE_TOLERANCE_MODE to M_ABSOLUTE. Color histograms are computed for each target area and defined color-sample, and a match score between each target area histogram and the color-sample histograms is generated. The color-sample with the highest score above the acceptance (McolControl() with M_ACCEPTANCE) is the target area's best-matched color-sample. Note that you cannot use M_HISTOGRAM_MATCHING if the distance type is M_MAHALANOBIS. (summarize) Specifies a minimum distance operation, based on color histograms. (more details...)
M_HISTOGRAM_VOTE	Specifies a match operation based on histogram voting. MIL computes color histograms for each defined color-sample and, for each target pixel, identifies its histogram bin. Each target pixel votes for the best color-sample, provided its histogram bin is not empty. The number of votes that a color-sample accumulates determines its score. The color-sample with the highest score above the acceptance (McolControl() with M_ACCEPTANCE) is the target area's best-matched color-sample. In the case of a tie, where more than one color-sample receives the same highest number of votes, the color-sample with the highest label value is chosen as the best-match. With histogram voting, you must set the distance type to M_NONE. (summarize) Specifies a match operation based on histogram voting. (more details...)
M_MIN_DIST_VOTE	Specifies a minimum distance operation, based on pixel voting. Color statistics are calculated (typically the mean) for each color-sample defined in the context, and the color distance between each pixel in each target area and each color-sample is determined. Each target pixel then votes for the closest color-sample, which is also within the distance tolerance (McolControl() with M_DISTANCE_TOLERANCE). The number of votes that a color-sample accumulates determines its score. The color-sample with the highest score above the acceptance (McolControl() with M_ACCEPTANCE) is the target area's best-matched color-sample. In the case of a tie, where more than one color-sample receives the same highest number of votes, the color-sample with the highest label value is chosen as the best-match. (summarize) Specifies a minimum distance operation, based on pixel voting. (more details...)
M_STAT_MIN_DIST	Specifies a minimum distance operation, based on pixel statistics. Color statistics are calculated (typically the mean) for each target area and for each color-sample defined in the context, and the color distance between each target area and each color-sample is determined. The resulting distances determine the score of the color-samples. If a distance is not within a color-sample's distance tolerance, this color-sample's score is 0%. The color-sample with the highest score above the acceptance (McolControl() with M_ACCEPTANCE) is the target area's best-matched color-sample. (summarize) Specifies a minimum distance operation, based on pixel statistics. (more details...)

For specifying the operation with which to establish the mapping that MIL associates between each color-sample and the reference color-sample, set this parameter to one of the following values. The ContextId parameter must indicate the identifier of a relative color calibration context.

For specifying the mapping operation (relative color calibration) INQ
Value	Description
M_DEFAULT	Same as M_HISTOGRAM_BASED.
M_COLOR_TO_COLOR	Specifies that a color-sample's mapping is based on transforming its color to the color of the reference color-sample on a point-to-point basis. It is recommended to have a pixel-wise correspondence between the color-sample and the reference color-sample. M_COLOR_TO_COLOR is the most precise operation, and the least flexible. Minor color variations and outlier color data can have a major impact. Use this operation when the conditions of your application are stable and you require an exact reproduction of color. (summarize) Specifies that a color-sample's mapping is based on transforming its color to the color of the reference color-sample on a point-to-point basis. (more details...)
M_GLOBAL_MEAN_VARIANCE	Specifies that a color-sample's mapping is based on transforming its mean and variance (standard deviation) to resemble the mean and variance of the reference color-sample. This is the fastest operation, and the least precise (most general). Use it when the color distribution of the reference color-sample and the color-samples is dissimilar, when conditions are not stable, or when other operations fail. M_GLOBAL_MEAN_VARIANCE can avoid the introduction of false colors if the acquisition scene (image and color data) varies dramatically between the reference color-sample and the color-samples. (summarize) Specifies that a color-sample's mapping is based on transforming its mean and variance (standard deviation) to resemble the mean and variance of the reference color-sample. (more details...)
M_HISTOGRAM_BASED	Specifies that a color-sample's mapping is based on using enhanced color distribution information to transform its color data to better resemble the color data of the reference color-sample. Use this operation when the color distribution between the reference color-sample and the color-samples is similar. M_HISTOGRAM_BASED is a compromise between point-to-point (high precision, low flexibility) and global mean variance (low precision, high flexibility) operations. It offers good precision without requiring a point-to-point correspondence. It also offers good flexibility, provided that colors are somewhat comparable. This operation is typically the longest to process. (summarize) Specifies that a color-sample's mapping is based on using enhanced color distribution information to transform its color data to better resemble the color data of the reference color-sample. (more details...)

DistTypeOrCalIntent INQ

Specifies the type of distance (for color matching) or color calibration intent (for relative color calibration) that the operation uses. Distance is the difference in color between the color-sample and the target area. Except for the M_MAHALANOBIS distance type, which uses the covariance of the color-sample, MIL uses the mean of the colors to calculate the distance. Color calibration intent is the extent to which the color data of the color-samples must resemble the color data of the reference color-sample.

See the Parameter associations section for possible values that can be specified.

ConversionModeOrComputeOption INQ

Specifies the conversion mode (for color matching) or compute option (for relative color calibration) that the operation uses. Conversion mode refers to the color space to which the RGB source color data will be converted before performing McolMatch(). You can only convert RGB source color data to CIELAB or HSL when matching colors. By default, this type of conversion is not performed (or allowed), and you should set this parameter to M_DEFAULT (same as M_NONE). Compute option indicates whether the operation uses all pixel values or color statistics.

When specifying a conversion mode, the match operation uses the converted color values. For example, if you are using the ConversionModeOrComputeOption parameter to convert your RGB color data to HSL, the match will be performed with the HSL version of the color.

See the Parameter associations section for possible values that can be specified.

ControlFlag

Reserved for future expansion and must be set to M_DEFAULT.

Parameter associations

The tables below list possible values for the ContextId, DistTypeOrCalIntent, and ConversionModeOrComputeOption parameters.

For specifying the type of distance calculation (color matching)
For specifying the color calibration intent (quality) of the mapping operation (relative color calibration)

The following ContextId, DistTypeOrCalIntent, and ConversionModeOrComputeOption parameter settings specify the type of distance calculation to use when calling McolMatch(). The ContextId parameter must indicate the identifier of a color matching context.

Except for converting RGB source color data to CIELAB or HSL when matching colors, the ConversionModeOrComputeOption parameter is unused and should be set to M_DEFAULT (same as M_NONE).

The settings in the following table are restricted according to the color matching context's source color space, as set using McolAlloc() with M_RGB, M_HSL, or M_CIELAB. Note that the only distance type available for an M_HSL source color space, as indicated in the table below, is M_MANHATTAN (which is also the default for HSL).

The M_CMC... and M_CIE... settings are recommended for advanced users and are typically applied to improve the Delta-E performance when dealing with minor color variances in industrial color difference evaluation. These distance types follow the standards of the International Commission on Illumination (CIE), as established in their technical report on Colorimetry (CIE 15:2004).

For specifying the type of distance calculation (color matching)
ContextId			Description
	DistTypeOrCalIntent
		ConversionModeOrComputeOption
CIELAB color matching context ID			Specifies a color matching context that is using an M_CIELAB source color space. (summarize) Specifies a color matching context that is using an M_CIELAB source color space. (more details...)
	M_DEFAULT		Specifies the default distance type. The default is the same as M_EUCLIDEAN. (summarize) Specifies the default distance type. (more details...)
	M_CIE94_GRAPHIC_ARTS		Specifies to use the CIE94 color distance with weighting factors pertaining to the graphic arts industry, as established by the CIE. The weighting factors are: KL = 1, K1 = 0.045, and K2 = 0.015. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIE94 color distance with weighting factors pertaining to the graphic arts industry, as established by the CIE. (more details...)
	M_CIE94_TEXTILE		Specifies to use the CIE94 color distance with weighting factors pertaining to the textile industry, as established by the CIE. The weighting factors are: KL = 2, K1 = 0.048, K2 = 0.014. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIE94 color distance with weighting factors pertaining to the textile industry, as established by the CIE. (more details...)
	M_CIEDE2000		Specifies to use the CIEDE2000 color distance, as established by the CIE. This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIEDE2000 color distance, as established by the CIE. (more details...)
	M_CMC_ACCEPTABILITY		Specifies to use the CMC(l: c) color distance that has been adjusted for acceptability evaluation, as established by the CIE. The adjusted parameters are: lightness (l) = 2 and chroma (c) = 1. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CMC(l: c) color distance that has been adjusted for acceptability evaluation, as established by the CIE. (more details...)
	M_CMC_PERCEPTIBILITY		Specifies to use the CMC(l: c) color distance that has been adjusted for perceptibility evaluation, as established by the CIE. The adjusted parameters are: lightness (l) = 1 and chroma (c) = 1. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CMC(l: c) color distance that has been adjusted for perceptibility evaluation, as established by the CIE. (more details...)
	M_DELTA_E		Specifies to use a Delta-E color distance, as defined by the International Commission on Illumination (CIE). A Delta-E color distance is the same as a Euclidean color distance (M_EUCLIDEAN), but specialized for CIELAB. (summarize) Specifies to use a Delta-E color distance, as defined by the International Commission on Illumination (CIE). (more details...)
	M_EUCLIDEAN		Specifies to use a Euclidean color distance. A Euclidean distance is the square root of the sum of the squared differences between the color of the color-sample and the color of the target area. (summarize) Specifies to use a Euclidean color distance. (more details...)
	M_MAHALANOBIS		Specifies to use a Mahalanobis color distance. A Mahalanobis distance is calculated between the color of the target area and, typically, the covariance of the color-sample. This distance, between a color and a distribution of colors, is similar to a Euclidean distance between the mean of the two colors, but weighted by the inverse of the covariance of the distribution. This implies that the more a color distribution varies in a direction within the color space, the less important is the distance in that direction. Since the covariance matrix of the color-sample is used, the color-sample should usually come from a an image, and not be defined as a triplet (McolDefine() with M_IMAGE instead of M_TRIPLET). However, if you provide a color constant (triplet) as the second source, Mahalanobis behaves very much like Euclidean and will yield similar results. M_MAHALANOBIS is not available if the operation mode is M_HISTOGRAM_MATCHING. (summarize) Specifies to use a Mahalanobis color distance. (more details...)
	M_MANHATTAN		Specifies to use a Manhattan color distance. A Manhattan distance is the sum of the absolute value of the differences between the color of the color-sample and the color of the target area. For an HSL color space, the distance between angular coordinates is equal to the smallest angular difference, rather than the absolute value of the difference. (summarize) Specifies to use a Manhattan color distance. (more details...)
	M_NONE		Specifies that no distance calculations are used. Instead, pixels in the target image vote for histogram bins. The histogram bins each have a range of values, and each pixel in the target image votes for the histogram bin whose range includes its value. M_NONE can only be set with M_HISTOGRAM_VOTE, and vice versa. (summarize) Specifies that no distance calculations are used. (more details...)
HSL color matching context ID			Specifies a color matching context that is using an M_HSL source color space. (summarize) Specifies a color matching context that is using an M_HSL source color space. (more details...)
	M_DEFAULT		Specifies the default distance type. The default is the same as M_MANHATTAN. (summarize) Specifies the default distance type. (more details...)
	M_MANHATTAN		Specifies to use a Manhattan color distance. A Manhattan distance is the sum of the absolute value of the differences between the color of the color-sample and the color of the target area. For an HSL color space, the distance between angular coordinates is equal to the smallest angular difference, rather than the absolute value of the difference. (summarize) Specifies to use a Manhattan color distance. (more details...)
	M_NONE		Specifies that no distance calculations are used. Instead, pixels in the target image vote for histogram bins. The histogram bins each have a range of values, and each pixel in the target image votes for the histogram bin whose range includes its value. M_NONE can only be set with M_HISTOGRAM_VOTE, and vice versa. (summarize) Specifies that no distance calculations are used. (more details...)
RGB color matching context ID			Specifies a color matching context that is using an M_RGB source color space. (summarize) Specifies a color matching context that is using an M_RGB source color space. (more details...)
	M_DEFAULT		Specifies the default distance type. The default is the same as M_EUCLIDEAN. (summarize) Specifies the default distance type. (more details...)
		M_DEFAULT	Same as M_NONE.
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation. When converting RGB source colors before the match using McolSetMethod() with the CIELAB conversion mode, MIL requires uncorrected (linear) color data. Therefore, if gamma correction has been applied on the RGB source color data, you must remove it using McolControl() with M_CONVERSION_GAMMA set to M_ENABLE. (summarize) Specifies to internally convert RGB colors to CIELAB before the match operation. (more details...)
		M_HSL	Specifies to internally convert RGB colors to HSL before the match operation.
		M_NONE	Specifies no internal conversion.
	M_CIE94_GRAPHIC_ARTS		Specifies to use the CIE94 color distance with weighting factors pertaining to the graphic arts industry, as established by the CIE. The weighting factors are: KL = 1, K1 = 0.045, and K2 = 0.015. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIE94 color distance with weighting factors pertaining to the graphic arts industry, as established by the CIE. (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_CIE94_TEXTILE		Specifies to use the CIE94 color distance with weighting factors pertaining to the textile industry, as established by the CIE. The weighting factors are: KL = 2, K1 = 0.048, K2 = 0.014. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIE94 color distance with weighting factors pertaining to the textile industry, as established by the CIE. (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_CIEDE2000		Specifies to use the CIEDE2000 color distance, as established by the CIE. This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CIEDE2000 color distance, as established by the CIE. (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_CMC_ACCEPTABILITY		Specifies to use the CMC(l: c) color distance that has been adjusted for acceptability evaluation, as established by the CIE. The adjusted parameters are: lightness (l) = 2 and chroma (c) = 1. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CMC(l: c) color distance that has been adjusted for acceptability evaluation, as established by the CIE. (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_CMC_PERCEPTIBILITY		Specifies to use the CMC(l: c) color distance that has been adjusted for perceptibility evaluation, as established by the CIE. The adjusted parameters are: lightness (l) = 1 and chroma (c) = 1. For more information, including the relevant formulas, refer to the technical report on Colorimetry (CIE 15:2004). This distance type does not support band selection; that is, McolControl() with M_BAND_MODE must be set to M_ALL_BANDS (or M_DEFAULT). (summarize) Specifies to use the CMC(l: c) color distance that has been adjusted for perceptibility evaluation, as established by the CIE. (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_DELTA_E		Specifies to use a Delta-E color distance, as defined by the International Commission on Illumination (CIE). A Delta-E color distance is the same as a Euclidean color distance (M_EUCLIDEAN), but specialized for CIELAB. (summarize) Specifies to use a Delta-E color distance, as defined by the International Commission on Illumination (CIE). (more details...)
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
	M_EUCLIDEAN		Specifies to use a Euclidean color distance. A Euclidean distance is the square root of the sum of the squared differences between the color of the color-sample and the color of the target area. (summarize) Specifies to use a Euclidean color distance. (more details...)
		M_DEFAULT	Same as M_NONE.
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation. When converting RGB source colors before the match using McolSetMethod() with the CIELAB conversion mode, MIL requires uncorrected (linear) color data. Therefore, if gamma correction has been applied on the RGB source color data, you must remove it using McolControl() with M_CONVERSION_GAMMA set to M_ENABLE. (summarize) Specifies to internally convert RGB colors to CIELAB before the match operation. (more details...)
		M_NONE	Specifies no internal conversion.
	M_MAHALANOBIS		Specifies to use a Mahalanobis color distance. A Mahalanobis distance is calculated between the color of the target area and, typically, the covariance of the color-sample. This distance, between a color and a distribution of colors, is similar to a Euclidean distance between the mean of the two colors, but weighted by the inverse of the covariance of the distribution. This implies that the more a color distribution varies in a direction within the color space, the less important is the distance in that direction. Since the covariance matrix of the color-sample is used, the color-sample should usually come from a an image, and not be defined as a triplet (McolDefine() with M_IMAGE instead of M_TRIPLET). However, if you provide a color constant (triplet) as the second source, Mahalanobis behaves very much like Euclidean and will yield similar results. M_MAHALANOBIS is not available if the operation mode is M_HISTOGRAM_MATCHING. (summarize) Specifies to use a Mahalanobis color distance. (more details...)
		M_DEFAULT	Same as M_NONE.
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation.
		M_NONE	Specifies no internal conversion.
	M_MANHATTAN		Specifies to use a Manhattan color distance. A Manhattan distance is the sum of the absolute value of the differences between the color of the color-sample and the color of the target area. For an HSL color space, the distance between angular coordinates is equal to the smallest angular difference, rather than the absolute value of the difference. (summarize) Specifies to use a Manhattan color distance. (more details...)
		M_DEFAULT	Same as M_NONE.
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation. When converting RGB source colors before the match using McolSetMethod() with the CIELAB conversion mode, MIL requires uncorrected (linear) color data. Therefore, if gamma correction has been applied on the RGB source color data, you must remove it using McolControl() with M_CONVERSION_GAMMA set to M_ENABLE. (summarize) Specifies to internally convert RGB colors to CIELAB before the match operation. (more details...)
		M_HSL	Specifies to internally convert RGB colors to HSL before the match operation.
		M_NONE	Specifies no internal conversion.
	M_NONE		Specifies that no distance calculations are used. Instead, pixels in the target image vote for histogram bins. The histogram bins each have a range of values, and each pixel in the target image votes for the histogram bin whose range includes its value. M_NONE can only be set with M_HISTOGRAM_VOTE, and vice versa. (summarize) Specifies that no distance calculations are used. (more details...)
		M_DEFAULT	Same as M_NONE.
		M_CIELAB	Specifies to internally convert RGB colors to CIELAB before the match operation. When converting RGB source colors before the match using McolSetMethod() with the CIELAB conversion mode, MIL requires uncorrected (linear) color data. Therefore, if gamma correction has been applied on the RGB source color data, you must remove it using McolControl() with M_CONVERSION_GAMMA set to M_ENABLE. (summarize) Specifies to internally convert RGB colors to CIELAB before the match operation. (more details...)
		M_HSL	Specifies to internally convert RGB colors to HSL before the match operation.
		M_NONE	Specifies no internal conversion.

The following ContextId, DistTypeOrCalIntent, and ConversionModeOrComputeOption parameter settings specify the color calibration intent (quality) with which to perform the operation that MIL uses to establish the mapping it associates to color-samples. These settings can increase or decrease the accuracy inherent in the operation. The ContextId parameter must indicate the identifier of a color matching context. Unless otherwise specified, values are available for all operations.

Except for converting RGB source color data to CIELAB or HSL when matching colors, the ConversionModeOrComputeOption parameter is unused and should be set to M_DEFAULT (same as M_NONE).

For specifying the color calibration intent (quality) of the mapping operation (relative color calibration)
ContextId			Description
	DistTypeOrCalIntent
		ConversionModeOrComputeOption
Relative color calibration context ID			Specifies a relative color calibration context. If using M_GLOBAL_MEAN_VARIANCE, set this parameter to M_DEFAULT. (summarize) Specifies a relative color calibration context. (more details...)
	M_DEFAULT		Specifies the default color calibration intent. For M_COLOR_TO_COLOR and M_HISTOGRAM_BASED, the default is M_BALANCE. For M_GLOBAL_MEAN_VARIANCE, the default is M_GENERALIZATION. (summarize) Specifies the default color calibration intent. (more details...)
		M_DEFAULT	Same as M_COMPUTE_ITEM_PIXELS.
		M_COMPUTE_ITEM_PIXELS	Specifies that MIL uses all pixel values (in the reference color-sample and the color-samples) to generate the color information that the operation requires.
		M_COMPUTE_ITEM_STAT	Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. You can only specify this value if using an M_COLOR_TO_COLOR operation. (summarize) Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. (more details...)
	M_BALANCE		Specifies that the operation considers a moderate amount of color information to establish the mapping. Though this setting offers less accuracy than M_PRECISION, there is little risk of overfitting the color data. You cannot specify M_BALANCE if you are using an M_GLOBAL_MEAN_VARIANCE operation. (summarize) Specifies that the operation considers a moderate amount of color information to establish the mapping. (more details...)
		M_DEFAULT	Same as M_COMPUTE_ITEM_PIXELS.
		M_COMPUTE_ITEM_PIXELS	Specifies that MIL uses all pixel values (in the reference color-sample and the color-samples) to generate the color information that the operation requires.
		M_COMPUTE_ITEM_STAT	Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. You can only specify this value if using an M_COLOR_TO_COLOR operation. (summarize) Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. (more details...)
	M_GENERALIZATION		Specifies that the operation considers a minimal amount of color information to establish the mapping. This setting offers the least precise color mapping. However, there is little risk of overfitting the color data, and it is the fastest calculation, which can prove useful for large images. (summarize) Specifies that the operation considers a minimal amount of color information to establish the mapping. (more details...)
		M_DEFAULT	Same as M_COMPUTE_ITEM_PIXELS.
		M_COMPUTE_ITEM_PIXELS	Specifies that MIL uses all pixel values (in the reference color-sample and the color-samples) to generate the color information that the operation requires.
		M_COMPUTE_ITEM_STAT	Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. You can only specify this value if using an M_COLOR_TO_COLOR operation. (summarize) Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. (more details...)
	M_PRECISION		Specifies that the operation considers a high amount of color information to establish the mapping. This setting offers the most precise color mapping between the reference color-sample and the color-samples. However, it can potentially overfit the color data. For example, if you perform the relative color calibration on grabbed images containing outlier colors that were not considered during preprocessing, incorrect color transformations can occur. You cannot specify M_PRECISION if you are using an M_GLOBAL_MEAN_VARIANCE operation. (summarize) Specifies that the operation considers a high amount of color information to establish the mapping. (more details...)
		M_DEFAULT	Same as M_COMPUTE_ITEM_PIXELS.
		M_COMPUTE_ITEM_PIXELS	Specifies that MIL uses all pixel values (in the reference color-sample and the color-samples) to generate the color information that the operation requires.
		M_COMPUTE_ITEM_STAT	Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. You can only specify this value if using an M_COLOR_TO_COLOR operation. (summarize) Specifies that MIL uses color statistics (in the reference color-sample and the color-samples) to generate the color information that the operation requires. (more details...)

Compilation information

Header	Include mil.h.
Library	Use mil.lib; milcolor.lib.
DLL	Requires mil.dll; milcolor.dll.