Regim stacking Software

[Bottletopburly]

I had tried Regim stacking software before with sort of mixed results @dave.hardy2(note you do need Java to run this program ) info is not abundant except for the PDF document by Andreas Rorig http://www.andreasroerig.de/content/regim/regim_eng.pdf  ,and a setup tutorial by guy on Startools forum https://forum.startools.org/viewtopic.php?f=7&t=1048  , so this week i had another go but colour was not  right ,bit more digging and realized probably due to the camera and being modded too and   not being supported something i think to do with debayering , so back to andreas and guys tutorials and seemed i needed Libraw to be installed and setup within regim ,now i can get around a PC but no expert by any means ,so libraw installed and after seven attempts of different settings finally i think i cracked it ,and the resulting file looks like it should do  so i will share the settings i have so far come up with in case anyone is looking how to get it to work properly, now i use dithered images and no Darks so loaded bias frames into Darks folder when loading subs  later i found for me this has to be done as the resulting file  does not look correct ,obviously if your using darks then its a matter of loading into darks ,there is no folder for bias but there is one for flats and flat darks .

so first thing to set is preferences, -File > preferences  here is where libraw files path is set , the files are in the bin folder within  libraw folder ,just a matter of navigating to the Path ,the two files you need to input are Raw.exe and Identity.exe 

 

Next tab to do is Preprocessing and >Files , on right hand side of screen there are some tick boxes ,just tick as screenshot 

Next is calibration > Preprocessing > calibration as per screen shot this is for my 1000D astro modified so choose Custom and tick first row is red pixel as per screenshot if unmodded then tick Standard DSLR  not sure if you have to choose first pixel is green may be a bit of trial and error using this setting and tick the other settings below as per screenshot regards to stretch raw to 16 bit , scale etc 

Then comes register >Preprocessing >Calibration  ,though you need to load up your subs to be able to tick the parameters box on right hand side of screen 

 

That's pretty much all you need to do settings wise just a matter of loading your subs lights and darks and flats and pressing Ok at the bottom of screen  in the preprocessing >file tab ,takes a while to do the only other thing i wasn't sure of was adding Data but Andreas has answered my question via the Regim yahoo group his answer is as follows

 

Hi Dave,

 

your settings look ok so far.

Concerning images from multiple nights:

If your conditions changed and/or you had to remove the camera so you needed a different set of flats, do calibration separately for each night (just select “do calibration” in the dialog).. Then copy all the calibrated files into one folder and run preprocessing without “do calibration” with the calibrated files as input..

 

Best regards

Andreas

 

I haven't tried this yet but will try once i get a minute , im by no means an expert but hopefully this may be of use to someone looking into Regim 

I tried on a image of M42 , regim result below followed by a DSS of same data both processed in Startools ,if you found this info useful and manage to stack a reasonable result please share , regim also has the ability to find the correct star colours but haven't looked into that part yet .

 

Regim 

DSS 

 

 

[+RonC]

That's quite a difference Dave, I will have to read through this tomorrow (when I'm more awake) thanks for all the info..?

[dave.hardy2]

Thanks Dave it looks promising 

[Bottletopburly]

Colour needs tweaking Andreas has responded to my question regards to getting preference parameters correct for my camera reply follows .

May be helpful for others .note I installed libraw 

 

what settings are needed also depends on the RAW converter you use in Regim (internal, or LibRaw or DCRaw, I recommend LibRaw). The best way to get the pixels right is to use a daylight photo of something that contains large red green and blue areas. Use custom and try all the combinations for the pixel order.

For a modified camera I would leave Auto scaling and Auto offset off, the scaling values all at 100 and the offset values 0.
This will result in an image with a strong red bias, but the automatic B-V color correction will take care of this.
 

Edited April 11, 2019 by Bottletopburly

[dave.hardy2]

Just a quick question. Can you stack images from different days? In DSS you use the tabs but I can't see anything on regim.

[Bottletopburly]

53 minutes ago, dave.hardy2 said:

Just a quick question. Can you stack images from different days? In DSS you use the tabs but I can't see anything on regim.

Yes you can ,

 

On 07/04/2019 at 20:07, Bottletopburly said:

Concerning images from multiple nights:

If your conditions changed and/or you had to remove the camera so you needed a different set of flats, do calibration separately for each night (just select “do calibration” in the dialog).. Then copy all the calibrated files into one folder and run preprocessing without “do calibration” with the calibrated files as input..

Andreas reply i haven't tried yet dave though seem to have cracked the debayer problem my 1000D is modded -i chose custom -first pixel red, 

 remove gradients tool very good  here is my latest go at m42 ,processed in Startools i will try and write the workflow down and post on this thread , very happy when you look at previous efforts and certainly more powerful than DSS

 

[dave.hardy2]

It is a learning curve but that's what we all love about this hobby isn't it

[dave.hardy2]

20 hours ago, Bottletopburly said:

Yes you can ,

 

Andreas reply i haven't tried yet dave though seem to have cracked the debayer problem my 1000D is modded -i chose custom -first pixel red, 

 remove gradients tool very good  here is my latest go at m42 ,processed in Startools i will try and write the workflow down and post on this thread , very happy when you look at previous efforts and certainly more powerful than DSS

 

That is a very good effort 

[Bottletopburly]

Regim workflow -M42  - assuming you have configured parameters needed in Regim first .

 

Processing tab >  click processing bottom of menu  > 

 

Using the browse to bar open the folder containing your light files, press add to Lights button on right hand side of screen.

Repeat actions for Flats ,Darks and Bias(flat Darks) files  ,(NOTE if using dithering add your bias files into Add to Darks)

 

Once all files loaded into relevant boxes press OK button at bottom of screen ,and let it process files this may take a while .

 

once processing finished Save file to a folder of your choosing so you can find easily > FILE >Save As >i save file as a FITS file as i process in startools  .

 

Close processing box once processed file has been saved .

 

Now press Edit tab > Remove gradient >  press Generate button > press execute 

 

leave main Image open >TOOLS tab >plate solving > Object name i am processing M42 so in object field  input messier 42 > i leave all fields as default and press OK 

image will now be platesolved ,leave until image solved .

 

Tools tab >Stars List> Press search  once all fields are populated with info  clos box 

and >File > save File as fits 

 

All info from plate solving and Stars list is now saved into image 

 

Keep image open ,now Color tab > automatic B-V calibration > press ok and let B-V calibration process ,once done Save file once again  

 

[dave.hardy2]

I think I have it this was last night 50 x 90secs with flats and bias frames no darks with a near full moon

Stacked in Regim Processed in Startools and CS6

 

M44 The Beehive Cluster

[Bottletopburly]

5 hours ago, dave.hardy2 said:

I think I have it this was last night 50 x 90secs with flats and bias frames no darks with a near full moon

Stacked in Regim Processed in Startools and CS6

 

M44 The Beehive Cluster

Excellent result Dave well done ? I imaged m13 last night so will see later how it went , I think I sussed multiple nights images just need to have a play in startools and see if I was successful, basically you stack in regim as normal save result for each separate night then you redo a stack but you only use the final images produced and untick do calibration this should add both images together then from this you can run gradient removal b-v colour etc, well that’s how I think it happens if I followed Andreas correctly, Andreas looked over my workflow and says all seems correct, have you tried the plate solving and gradient remover Dave ?

[dave.hardy2]

1 hour ago, Bottletopburly said:

Excellent result Dave well done ? I imaged m13 last night so will see later how it went , I think I sussed multiple nights images just need to have a play in startools and see if I was successful, basically you stack in regim as normal save result for each separate night then you redo a stack but you only use the final images produced and untick do calibration this should add both images together then from this you can run gradient removal b-v colour etc, well that’s how I think it happens if I followed Andreas correctly, Andreas looked over my workflow and says all seems correct, have you tried the plate solving and gradient remover Dave ?

I have now. Here is my latest just reprocessed, my last one didn't get the gradient to work but I have now sorted it I will have a go at multiple nights later.

 

[dave.hardy2]

I have combined 10 x 300sec with the 49 x 90sec in Regim and it worked a treat I did it just how you suggested. I do have an issue with the gradient removal though, when it finishes it flashes up with an image with the gradient removed then it just goes back to the image with the original gradient so not sure what is going on with that.

 

[Bottletopburly]

6 hours ago, dave.hardy2 said:

I have combined 10 x 300sec with the 49 x 90sec in Regim and it worked a treat I did it just how you suggested. I do have an issue with the gradient removal though, when it finishes it flashes up with an image with the gradient removed then it just goes back to the image with the original gradient so not sure what is going on with that.

 

You need to save it ,if you look at the bottom header it will say not saved ,once saved disappears 

[Bottletopburly]

Just now, Bottletopburly said:

You need to save it ,if you look at the bottom header it will say not saved ,once saved disappears 

Actually looks lot better  with twice data ?

[dave.hardy2]

Tonight I've gone for NGC2420 The Twinkling Comet Cluster it looks nice.