Primaries

The Primaries transformation is a way to alter color images by redefining the primary colors upon which its RGB color space is based. The ICC profile of the input image defines its white point and red, green and blue primary colors in terms of their CIE chromaticity coordinates. These coordinates are a device-independent color representation which let the color management system to convert the image to other color spaces.

This transformation lets you modify the white point and RGB primaries by first converting the image to device-indepenent CIE XYZ coordinates based on the input image primaries and then converting from CIE XYZ back to RGB based on a new white point and RGB primaries.

At the bottom of the Primaries dialog box is a standard CIE diagram overlayed with a triangle and a point in the middle . The vertices of the triangle mark the locations of the input image primaries and the point in the middle represents the input image white point.

To alter the white point or primaries, click and drag them to define the output image primaries and white point. As you alter them, a line is displayed connecting the input values (which remain fixed) to the output values.

Above the CIE diagram, the numeric input and output chromaticity coordinates are displayed for reference.

Amount

The amount control lets you control how much of the precise Primaries transformation is applied to the input image. You can apply a percentage of the transformation to the entire image, or you can specify an amount mask to restrict the effects of the transformation to only part of the input image.

Tips

Moving a primary towards the center decreases the amount of that color in the image and increases the amount of the color on the opposite side of the white point. For example, moving the blue primary towards the center make the image less blue and more yellow.

Small changes in the white point make a big difference. If necessary, you can resize the dialog box to make the diagram easier to edit.

Especially if you make large changes, the results can be hard to predict and it is likely that one or more of the RGB channels in the output image may be clipped. To check for clipped channels, you can enable clipped pixel display from the main tool bar:

Because the starting white point and primaries are derived from the input image color profile, applying the same transformation to another image with a different profile may not have exactly the same effect.