Stacking Canon RAW CR2 Files For Solar and Lunar Imaging

I use Deep Sky Stacker for stacking images from my 450D for wide field shots of the night sky, but I’ve struggled getting anything to work for solar and lunar images. Some people I’ve discussed it with have found no problems whilst others just haven’t been able to get it to work at all.

I was in the latter category and resorted to stacking the JPEG files produced by the camera rather than working from the RAW CR2 files which would always be my preference as the JPEG compression is lossy.

Initially I experimented with converting the CR2 files to TIFF using ImageMagick, but despite trying to stack those files with both versions 5 and 6 of Registax the stacking was very poor, and AutoStakkert 2 ran for ten hours without producing a usable result. Looking at the content of the TIFF files it appears that they contained a large amount of noise and I believe that to be the cause of the problem.

I then tried Canon’s DPP software which came with the camera. That can convert RAW images to TIFF in batch mode and appears to produce much cleaner files than ImageMagick. Again however all three stacking applications struggled to get anywhere, producing poor results.

Finally the author of PIPP suggested I try the new “almost-ready-for-release” version. It should read in the RAW files, crop them to a box just larger than the image and save them again as TIFF. I had a few niggles along the way but we got it to work fairly quickly and the resulting files now stack quite happily.

I intend to repeat this experiment when I next have the opportunity to do some solar imaging — the forecast is not looking good this week, but here are my results thus far. The processing is as far as possible the same except for the stacking, and all four images are taken from the same original imaging run. In the case of Registax, 54 frames were stacked in each case. I don’t know how many AS!2 chose to stack.

First, the result from using the JPEG files:

Then from the RAW files, preprocessed with PIPP and stacked with Registax v5:

And the same files stacked with Registax v6:

And finally stacked with AutoStakkert 2:

I think it’s a close call between the last two to decide which is best. I believe I can see a purple-ish “ghost” above the large dark sunspot towards the top left of the Sun in the last image though, so for the moment, for me, I think the third comes out marginally on top.

This entry was posted in Astroimaging and tagged , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *