The combination menu
Selecting rows in the table of currently opened images enables images to be combined in a variety of different ways.
- is only available if at least 2 rows have been selected. The selected images are added together, pixel by pixel, channel by channel.
- is only available if exactly 2 rows have been selected. A dialogue appears that shows a preview area from the centre of the image and a slider beneath it. At first the slider is central and the preview image shows equal amounts of the 2 images added together. When the slider is moved to the left it alters the proportion so that more of image 1 is used than of image 2. Fully to the left, the result is simply image 1. Sliding to the right uses more of image 2. The preview area can be dragged around to see the effects on other parts of the image. Pressing the OK button applies the proportional addition to the whole of the images, putting the result in image 1.
- is only available if at least 2 rows have been selected. The function is very similar to image addition except that the pixel values are divided by the number of images at the end.
- is only available if exactly 2 rows have been selected. The image in the second selected row is subtracted from the one in the first selected row, pixel by pixel, channel by channel. Subtraction can produce negative values so the result is scaled back into the range of levels available for the number of bits (0..255 for 8-bit channels, 0..65535 for 16-bit channels). Therefore pixels which are unchanged between the two images appear as mid-grey in the result. Pixels which were brighter in the first image than in the second are brighter than mid-grey in the result. Pixels which were darker in the first image are darker than mid-grey in the result.
- is also only available if exactly 2 rows have been selected. The only difference from the previous option is that the images are subtracted in the opposite order: the image in the first selected row is subtracted from the later one.
- is also only available for a pair of images. The later image in the list is multiplied into the earlier one, pixel by pixel. If M is the maximum level which can occur in each channel of the target image and N is the maximum channel level for the other image being multiplied, the result for each pixel is scaled linearly from M * N back to M.
- is only available if exactly 2 rows have been selected. A dialogue appears that shows a preview area from the centre of the image and a slider beneath it. At first the slider is central and the preview image shows equal amounts of the 2 images multiplied together. When the slider is moved to the left it alters the proportion so that more of image 1 is used than of image 2. Fully to the left, the result is simply image 1. Sliding to the right uses more of image 2. The preview area can be dragged around to see the effects on other parts of the image. Pressing the OK button applies the proportional multiplication to the whole of the images, putting the result in image 1.
- is also only available when a pair of images has ben selected. In astrophotography this can be a very useful thing to do, to divide by a "flat field" which removes instrumental variation across the image. For more information about the meaning and purpose of a flat field image, see here. True division of images is problematical because zero pixels in the dividend image produce an infinite result. For now, try multiplying by an inverted image instead. GRIP also offers another operation for removing background variation: background correction. Inversion and background correction are available on the levels menu of each image window.
-
is available only when 3 images are selected in the table. It swaps channels between the images like this:
Image 1: R1 G1 B1 becomes R1 G2 B3
Image 2: R2 G2 B2 becomes R2 G3 B1
Image 3: R3 G3 B3 becomes R3 G1 B2
This operation can be useful for testing and for recombining images which have been split into separate channels for different processing on each channel. It can also produce some interesting creative effects, especially for moving subjects. Applied 3 times it recreates the orginal 3 images.
Here is a creative use of this operation, on 3 successive images of sea waves breaking on rocks:
- is available if exactly 2 images have been selected in the table. An image is constructed in which the red channel is a monochrome version of the first image, the green channel is a monochrome version of the second image, and the blue channel is zero. The result can be viewed through red-green spectacles to see a three dimensional view. This works very well if you take two photos of a scene with exactly the same camera settings but move the camera by, say, 6 inches (15cm) between shots. The spectacles are readily available very cheaply from many firms that trade over the internet.
- is currently experimental and may not work well. The idea is to map one image onto another by interactively marking corresponding points. More information will be given when this is more robust.
- displays the present page if you have installed files as described here.