There are many software packages that can be used to process CCD
images. You can use just one package to process an image but is
often best to use several packages so you can combine the best features
of each one.
Processing Deep Sky Images
Calibrate, Align, Normalize and Combine the images
taken through each filter
Calibrate the images in
CCD Stack by subtracting
dark frames and
applying flat fields.
Align the images in
Registar. I have
tried the registration algorithms in MaximDL,
Images Plus and CCDStack
but find Registar to be the most accurate and reliable for
registering images. It is also extremely good at aligning
images that only have a small amount of overlap which is great for
Normalize the images in CCD Stack. This
makes each image the same relative brightness. It is important
that the images are normalized if the statistical noise rejection is
to work correctly.
Reject outlier pixels by using Standard Deviation
Sigma rejection in CCDstack. This works best when you have a
least 16 images. It will reject all pixels that fall a set
number of standard deviations from the average pixel value.
This technique is very effective at removing satellite an asteroid
trails as well as hot/cold pixels if the images were taken using
dithering. I set
CCD Stack to reject pixels greater than 2.2 standard deviations from
Combine the images in CCD Stack using Average and
save the master image
Final processing of the
luminance image in CCD Stack
Perform deconvolution if required to sharpen the
fine detail. This tends to work best to bring out the fine
detail in galaxies.
Perform a mild digital development stretch in CCD
Stack. This brings out the faint detail and avoids saturating
the bright areas.
Make the master color image from the master images
taken through the RGB filters
Use color combine in CCD Stack or Images Plus to make
a single color image out the master images taken through the red, green
and blue filters. You should adjust the ratio of each color based
G2V calibration for your camera/filter combination.
Create the LRGB
image in Photoshop
Align the color image with the luminance image in
Load the color image in
Photoshop and use levels to
stretch the image and make it brighter. Then look at the
histogram for each color and adjust the level of the black point so
that each histogram starts to rise at the same point. This
will ensure a correct color balance in your image.
Use levels and curves on the color image to
brighten it and bring out the fine detail.
is a great plugin for Photoshop that makes it very simple to remove
gradients from the color and luminance images.
Load the luminance image. Finish processing
it with levels/curves, sharpening, noise reduction, gradient removal
Copy the luminance image as a layer on top of the
color image. Set the combine mode for the layer to Luminance
and now you will have the fine detail of the luminance image with
the color of the color image shining through.
Increase color saturation of the color image by
applying the Shadows/Highlight filter and boosting the color
Blur the color image to reduce noise. A 3
pixel Gaussian blur works well.
Flatten the image and switch to Lab mode.
Select the 'a' channel and increase the contrast and then do the
same with the 'b' channel. This increases the saturation of
the mage without introducing noise.
Switch the image back to RGB mode and do any last
processing to taste.
There are many techniques you can use in Photoshop to
enhance your image, far more than I can list here. I recommend
getting a copy of Adam Block's
'Powerful Processing in Photoshop' DVD. All the techniques you
will need to produce great astro images can be found in a series of easy
to follow tutorials. Adam also has a great
DVD tutorial on
CCD Stack as well.
Starizona has a good online guide to
CCD imaging and
Processing Planetary Images
For planetary imaging with the webcam it is best to use
followed by Photoshop. Registax
is freeware software which can process AVI files captured by webcams.
A planetary imager may capture an AVI file of a minute's duration or more. It
will contain 100s or 1000s of images of the planet. Registax can
sort through these images, automatically picking the ones with the most
detail . It can then sum these together to produce a single image with a
much higher signal to noise ratio. Registax also contains some useful image
processing routines that can be used to enhance the detail in the summed
image. For example it contains a very powerful unsharp mask