9.2 Stretching the Curve
143
Color and Color Mode to RGB Color. Select the new image and the channel tab. You
will see a list of four images, RGB, Red, Green, and Blue.
Select the red channel and then select the red image. Select the whole of the red
image with a ctrl-A and then copy it with a ctrl-C (or just use the Edit drop-down
menu). Now return to the new image and select paste in place from the Edit menu.
Do the same for the green and blue channels until you have your colour image. You
now need to add the luminance frame. Select the Layers tab, then Layer New Layer.
Set the new layer mode to Luminosity. This will bring up a second layer above your
RGB image (which will be called Background). Select the new layer and then select
your luminance frame. Select all and then copy, then return to the new image and
paste in place. You now have an LRGB image.
The process is somewhat similar in GIMP. Open your three RGB frames as we
would in Photoshop, except we do not need to open the luminance frame yet. Also,
we do not need a fifth frame. From Colors select component and then compose.
Choose RGB as the Color Model and select the three frames corresponding to the
three filter colours; pressing OK results in a colour image. To add the luminance
frame, select the colour image and then select the luminance file from File Open as
Layers. Highlight the new Luminance layer in the Layer bar and change its mode to
value. Now left click on the luminance layer and select merge down. You now have
an LRGB image.
9.2.2 Image Filtering
Now that you have your colour image, you may wish to enhance it in order to bring out
the finer detail. This can be done by applying any number of mathematical processes
to the image, a process known as, somewhat confusingly, filtering. Filtering an image
is fundamentally different from stretching an image. Filtering is nonlinear, and once
accepted, it is permanent. It does not change how the image is displayed, but it
changes the pixel values. Hence, once you filter an image, you can no longer use it
for the purposes of science. However, for completeness, we will detail a number of
the more common image processing filters.
• High Pass We can consider the data in an image to be a series of waves representing
the rate of change between pixels. If the values change rapidly between pixels,
the data is of high frequency. The high pass filter accentuates the high-frequency
regions of the image, enhancing those areas at the cost of increased noise.
• Unsharp Mask The unsharp mask performs edge detection and enhances the
edges. In most implementations, you need to define the width of the edges, the
aggressiveness of the process, and the floor at which a pixel is considered part of
the background.
• Wavelets The wavelet process identifies specific frequencies (as selected by the
user) within the image, and either enhances or suppresses them. The process is
very powerful, but it requires considerable skill to use.
143
Color and Color Mode to RGB Color. Select the new image and the channel tab. You
will see a list of four images, RGB, Red, Green, and Blue.
Select the red channel and then select the red image. Select the whole of the red
image with a ctrl-A and then copy it with a ctrl-C (or just use the Edit drop-down
menu). Now return to the new image and select paste in place from the Edit menu.
Do the same for the green and blue channels until you have your colour image. You
now need to add the luminance frame. Select the Layers tab, then Layer New Layer.
Set the new layer mode to Luminosity. This will bring up a second layer above your
RGB image (which will be called Background). Select the new layer and then select
your luminance frame. Select all and then copy, then return to the new image and
paste in place. You now have an LRGB image.
The process is somewhat similar in GIMP. Open your three RGB frames as we
would in Photoshop, except we do not need to open the luminance frame yet. Also,
we do not need a fifth frame. From Colors select component and then compose.
Choose RGB as the Color Model and select the three frames corresponding to the
three filter colours; pressing OK results in a colour image. To add the luminance
frame, select the colour image and then select the luminance file from File Open as
Layers. Highlight the new Luminance layer in the Layer bar and change its mode to
value. Now left click on the luminance layer and select merge down. You now have
an LRGB image.
9.2.2 Image Filtering
Now that you have your colour image, you may wish to enhance it in order to bring out
the finer detail. This can be done by applying any number of mathematical processes
to the image, a process known as, somewhat confusingly, filtering. Filtering an image
is fundamentally different from stretching an image. Filtering is nonlinear, and once
accepted, it is permanent. It does not change how the image is displayed, but it
changes the pixel values. Hence, once you filter an image, you can no longer use it
for the purposes of science. However, for completeness, we will detail a number of
the more common image processing filters.
• High Pass We can consider the data in an image to be a series of waves representing
the rate of change between pixels. If the values change rapidly between pixels,
the data is of high frequency. The high pass filter accentuates the high-frequency
regions of the image, enhancing those areas at the cost of increased noise.
• Unsharp Mask The unsharp mask performs edge detection and enhances the
edges. In most implementations, you need to define the width of the edges, the
aggressiveness of the process, and the floor at which a pixel is considered part of
the background.
• Wavelets The wavelet process identifies specific frequencies (as selected by the
user) within the image, and either enhances or suppresses them. The process is
very powerful, but it requires considerable skill to use.