Need some advice..

[Tinbeard]

Hi,
I finally managed to get some images of M42 and I'm trying to use the HDR Composition tool but not having much luck, I'm not too sure if it's my images or not.
The issue is in the original images I have a strong colour cast and quite a few star halos and when I run the HDR tool I basically get the stars replaced with what looks like doughnuts.
I had a few integrated sets I'm trying to run in the HDR, these are:

10x 10min
10x 8min
10x 5min
10x 1min
10x 30sec
10x 10sec
10x 2sec

Is this too many ?

I do get an initial error when running the process about bad fitting FIT (or something similar) if I reduce the binarise setting this clears.

I've attached a couple of images one is the result of the HDR process using just luminance and stretch with STF, the other is a section of the 10min sub.

Edit: the 3rd image is the HDR result using the process described in my reply below.

Any help or advise would be greatly appreciated.

Edit: added another image of result with low binarise and quick histogram stretch.

Marc.

[Juan Conejero]

Hi Marc,

That sequence of exposure times seems reasonable for M42. The HDRComposition tool requires linear data. Are you trying to use nonlinear (stretched) images?

[Tinbeard]

Hi,

The steps I completed before using the HDRComposition tool is as follows:

• Calibrate each set of subs with master dark, bias and flat
• Register each set of subs
• Integrate each set of subs
• Run the HDRComposition tool with the integrated images

I'm not sure why the stars end up hollow, it seems like the HDRComplation tool replaced the actual stars with data from the 2 second exposure but left the star halos in place.

I'm very new to Pixinsight and image processing in general so I may be missing something very obvious.

Marc

[Cleon_Wells]

Marc, could you show us 1 of your 10 min _c.fit subs, STF  stretched and show the Statistics on the sub. Statistics is under Process/Image.
I've attached two screen shoots combined in PS of two different sub Statistics, before and after Calibration, note the Minimum levels of zero in the image, possibly caused by errors in my Calibration files.
Cleon

[Tinbeard]

Hi Cleon,

I've attached a screenshot of a calibrated 10min sub and I have minimum values of 0, is this down to poor dark, flat masters ?

I'll try and do the whole process again but this time I won't calibrate the frames just to see if the donught effect disappears.

Marc.

[Cleon_Wells]

Marc, after you calibrated this sub did you DeBayer the _c.fit file? (Scripts/BatchProcessing)
When you use the STF function remove the highlight square around the RGB blocks (in the upper left corner) and then click the A button.
Cleon

[Tinbeard]

Hi,

I have repeated the screen shot with the RGB highlight off in STF.

The original subs have not been debayered as they were not RAW captures ( My understanding is only RAW needs this process, but correct me if I'm wrong).

To be on the safe side the screenshot is of a DeBayered calibrated sub with the associated stats.

Thanks for taking the time to assist me with this.

Marc.

[Cleon_Wells]

Marc, I think I can help but  no guaranty.

The original subs have not been debayered as they were not RAW captures ( My understanding is only RAW needs this process, but correct me if I'm wrong).

If the images are taking with a DSLR/Bayer matrix/cfa camera and convert the CR2/raw files to a .fit format you have to DeBayer them  after you calibrate them. Attached are some important settings for processing. Also skim over this. http://www.stark-labs.com/craig/articles/assets/Debayering_API.pdf
Cleon

[Tinbeard]

Hi,

These were all captured using Nebulosity 2 and the camera was set to jpg so I'm guessing the camera takes care of the Debayering of the images. I will start to use Cr2 but don't want any additional complexity until I get a bit more proficient in image processing.

I repeated the HDRComposition but this time just using a single sub from each set that has not been calibrated with the bias,dark and flat. The results look an awful lot better that the initial attempt with the integrated sets of subs that had been calibrated. The stars no longer look like rings so this is a major step forwards 

Do you think it's a claibration frame issue ?

Marc



 

[Cleon_Wells]

Marc, I have never tried .jpg files so I am not sure what should/will happen.
Cleon

[georg.viehoever]

Setting the camera to JPG is not a good idea: The images loose a lot of depth, plus the camera does some kind of non-linear transformation. I doubt that it is possible to produce good HDR results with this.
Georg

[Tinbeard]

Ah well, Ill give the HDRComposition a miss for now and capture some RAW images before I try again.

Thanks for all your help and advise.

Marc