This is an experiment, to use a mono camera plus filters to image the moon. The result: Don't do that.
It turns out aligning images sufficiently well so that there are no color halos is very difficult. I tried numerous software with mixed results. This example appeared the best, it used Pixinsight to do an alignment using the FFTREgistration script, then a regular integration, color combination, and generation of a synthetic luminance from the mean of the three channels.
The resulting detail was not great, there are still alignment issues if you look closely. It also used all images in the stack, it did not do a quality assessment as I was unsure how I could do one with this approach (other than by eye I guess).
This might be compared to the mono image I will post on the same day from the same data, which uses 40% of the best images but treats them all as mono, and stacks in AutoStakkert.
I welcome suggestions and criticism, especially if you have a better way to do alignment of separate R, G, B data.
Linwood
Comments