Combine Color

The Combine Color command is used for combining up to four monochrome images into a single color image. Many cameras require that monochrome images be taken through various color filters and then combined afterwards into a color composite image.

This command can be used to compare two or three unfiltered images by producing a composite and looking for color differences. Comparisons can also be done using the Animate command.

The first step in color conversion is to select the Conversion Type. Six options are available:

When using LRGB or LCMY, select the desired Conversion color space. You may pick CIE L*a*b* (also referred to as CIELAB), RGB, or HSL (Hue-Saturation-Lightness). For CIELAB and HSL, an image synthesized from just the color data is converted to the specified color model, then the L data is replaced with the Luminance image and the result reconverted to tricolor. When RGB is selected, the output color image is produced by scaling the luminance image at each pixel by the ratios between the color data. The HSL algorithm is courtesy John Winfield.

The second step is to select the images to be combined. The image which was active prior to invoking the command will appear in one of the slots, as determined by examining the first letter of the value of the FILTER keyword in the FITS header, or in the absence of a FILTER keyword, the last letter of the file name. In RGB and LRGB modes the letters R, G, B are taken to denote Red, Green, and Blue images respectively, while in CMY and LCMY modes C, M, Y denote Cyan, Magenta, and Yellow. When using the FILTER keyword, any other letter is taken as luminance(L); when parsing frame information from the filename, only L or W (white) is accepted for this. Using the same identification algorithm, the command automatically inserts into the other dropdowns any other open images that were assigned the appropriate filters and whose filenames match the key image name. [When the FILTER keyword method is used, an approximate match is accepted.]   All images are entered in every control, so you can use the drop list to correct a wrong selection.

When using LRGB or LCMY combine modes, the images are automatically resized to a common binning factor used, if needed. The resultant binning will be that of the image with the smallest binning, which is usually the luminance frame. The scale factors are determined from the XBINNING and YBINNING keywords in the FITS header.  Combine Color normally uses the Double Size command in order to achieve the best final results when the destination binning is half that of the source (for example, converting 2x2-binned images to match an unbinned luminance image). If you prefer to use the equivalent of a straight Resize command, check Fast Resize. This control is not enabled unless it is applicable to the particular images being combined; when none of the selected images have binning ratios of 2:1, Fast Resize is always implied. Resizing for Preview Images also always uses the fast method.

If the images are not aligned prior to executing this command, this will be obvious in the Preview Image, provided Auto Preview is on (leaving this on is recommended). Click Align to invoke the Align dialog and register them. The first image (Red, Cyan, or Luminance, depending on the combine mode) is automatically selected as the reference to which the others are adjusted. Changes made in Align mode modify the selected image buffers, even if Combine Color is subsequently canceled, but can be rolled back via Undo if necessary. Control will automatically return to the Combine Color command when the Align Images dialog box is closed. Performing an alignment restricts the channel selection droplists to the set of images that were aligned. Also, you cannot perform an alignment more than once in the same Combine Color activation.

The next step is to set the balancing adjustments (Input / Output matrix). For standard RGB imaging, there are only three slots in the matrix available – Red Input / Red Output, Green Input / Green Output, and Blue Input / Blue Output. If the exposures in the three color bands are equal and the transmittance of the filters are comparable, these can all be set to 1 (click Default for this setting). If the exposure level in the blue band, for example, is half that in the other two, it should be set to 2 and the others to 1. The simplest way to determine the correct settings is to image a standard gray card using the same camera and optical system used for imaging. Adjust the exposures or the scaling so that the resulting image looks gray.

The balancing is more complex in CMY mode because it is a subtractive method. The default setting (obtained by clicking Default) corresponds to:

Red = Magenta + Yellow - Cyan

Green = Cyan + Yellow - Magenta

Blue = Cyan + Magenta - Yellow

CMY imaging is a more difficult technique which requires some effort in order to obtain correctly calibrated results. The simplest method is to take an image of a gray card, and adjust the exposures or scaling so that the resulting image looks gray. The simplest scaling adjustment is to use the same absolute value for each row in the table (keeping the additive/subtractive recipe as shown above). This is equivalent to scaling the exposure duration. More complex transformations are possible using different numbers in all nine slots.

If a Luminance plane is selected (LRGB or LCMY) the color information is taken from the three color images, and the luminance (brightness) level for each pixel is taken from the Luminance image. The human eye is less sensitive to color variations than brightness variations. This opens up the possibility of combining lower-resolution, short exposure color images with a single longer exposure for luminance. When Lum Wt % (Luminance Weight) is set to 100%, the luminance value derived from the color planes is completely replaced by the luminance frame; smaller values blend the luminance frame with information from the three color frames. You can see the effect of the Luminance information by switching between Conversion Types (e.g. LRGB to RGB) with Auto Preview enabled.

The Bgd Auto Equalize check box turns on automatic background equalization. It adjusts the images to compensate for any differences in the average background level. In most cases this should be turned on to equalize the sky background. In some cases it may be necessary to turn this off and make any necessary adjustments later in the Color Balance command. An example is an image which is completely full of dense nebulosity – there is no background in the image.