M42: What is the best way to combine 5 different exposures?

[markcasazza]

I recently shot M42 at 60 seconds, 30 seconds, and 10 seconds. I processed the 60 second stacked image two different ways. One was overly aggressive to get the faint nebulosity and is somewhat afflicted with noise. I also processed the 10 second frames two different ways. One normal and with much less stretch to retail even more of the very bright detail. I am now looking for the best way to combine all 5 "exposures." I watched the M42 demo video on YouTube, but it only used 2 exposures and I am asking, among other things, should I start with the dimmer and add brighter images (or the other way around) and how do I best deal with the noise on the aggressive 60 second exposure? I'm planning a masked application of a de-noise algorithm once they are all combined, but I do need to limit the impact of the noise when I layer them where I don't "need" the noisy data. I have place the reduced JPG versions of the 5 files for your review at http://casazza.net/M42

Thanks,

Mark

[admin]

Hi Mark,

Combining multiple exposure lengths is rarely needed or, indeed, a good idea. Save from the very brightest objects such as M42, objects are simply too faint to quickly overwhelm your sensor.

Mixing multiple exposure lengths for your primary data may at best lead to inconsistencies in noise levels and at worst lead to artefacts. For this reason, the HDR tutorial/video shows how to combine two 'finished' images.

You should/will get much better results by processing the 10 and 60, or 30 and 60 second stacks once and combining those. Stretch the maximum you want and let the Contrast, HDR, Sharp and Denoise modules do the rest.

If you really want to combine 5, however, you should be able to to simply keep repeating the process, constantly adding the next exposure to the output of the combination of the previous two. Rinse, repeat.

Hope this help,

[markcasazza]

Ivo,

Sorry I was not clear. I took 65 60-second, 43 30-second, and 33 10-second exposures all at ISO800. I stacked each set separately in Nebulosity. I then processed each set separately in StarTools. I used my StarTools workflow on every stack. For the 60 and 10 second stack I used the ST workflow twice with different "intensities". Now that I have 5 composite sub-frames I am looking to HDR combine them, I did it with PhotoShop tonight, but I am still interested in what StarTools could do. My end result is at: http://casazza.net/Gallery/displayimage ... fullsize=1

Mark

[Burly]

Best to process in DSS each sub set is on a separate tab, lights darks flats bias , Dss will then process all to give a combined image .

[admin]

Hi Mark,

If you'd like to use the HDR combine as seen in the video on more than two images, as I was trying to say, simply keep repeating the process, constantly adding the next exposure to the output of the combination of the previous two. Rinse, repeat.

Let's for example say we have stack of 10 seconds, 30 seconds, 60 seconds and 120 seconds.

Using the method in the video, combine 10 and 30.
Then use the same method to combine the output of the above with the 60 second stack.
Then use the same method to combine the output of the above with the 120 second stack.

Does that make sense?

Or like Burly says, DSS can combine multiple exposure lengths as well during stacking, however this will have ramifications for noise levels (and point spread functions) in your data which will vary all over the place.