Filter

This transformation simulates viewing the current image through a colored filter. The colored filter can either be a solid color or it can be any image whose width and height in pixels are the same as those of the input image. In either case the result is like sandwiching two transparencies and photographing the result.

Amount

The amount control lets you control how much of the filter 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.

Filter

This control lets you select the filter color or image. If you select an image, it must have the same dimensions in pixels as the input image.

Complement

Clicking this checkbox causes the filter color or color image selected in the Filter control to work as though it were replaced by the complementary color. Thus, to remove red from an image, instead of having to select a cyan filter (cyan being the complement of red), you can simply select a red filter and then click on the Complement checkbox. The more saturated the red color you select, the more red will be removed from the input image.

Method

This control lets you select what type of filtering you want to perform:

Additive -- filtering produces the effect of sandwiching the input image with the filter color or image and viewing them on a light box. The net result is that wherever the input image is white the result image takes on the filter color and wherever the input image is black, the result image is also black. Mathematically, filtering is done by multiplying the two images together.

Subtractive -- filtering is analogous to sandwiching two negatives together and then taking a negative of the result. The net result is that wherever the input image is black the result image takes on the filter color and wherever the input image is white, the result image is also white. Mathematically, subtractive filtering is done by inverting both the input image and the filter, multiplying them together, and then inverting the result.

Exposure Compensation

This slider lets you compensate for the tendency of some filters to darken the input image. The larger the compensation factor you choose, the brighter the result image will be. The slider is calibrated in f stops; in other words, each increase of 1.0 on the slider readout corresponds to brightening the image by a factor of two.

To check for highlight clipping, turn on Show Clipped Highlights  in the main tool bar.

Tips

You can use one of the Gradient transformations to create a custom graduated neutral density or color filter image the same size as the input image.

Filtering an image with itself makes it darker and more saturated.