12 replies to this topic

[Xshovelfighter]

Hey guys, is it acceptable/useful to stack and/or process images with differing exposure times? I've got two sets of image data for m51 that were collected on two separate nights. I had no intentions of combining the data from multiple nights so ended up using slightly different exposure times based on the seeing conditions.

However, we've got some cloud cover now and I was just messing around with Photoshop. One night was 50 second exposures and the other were 70 seconds. I just figured combining the data would basically double my total exposure time - but just wanted to see if it would be of any benefit.

My main concern is with the logistics of actually stacking the data. Or would I simply combine the layers in Photoshop and stack each set separately? I would like to combine everything in DSS but am not sure how that would work with the calibration frames and different exposure times. Any advice is appreciated!

[xiando]

I've had mixed success grouping different exposure times in DSS. I prefer to pre-process (register,align,stack)  them separately and then mix the results using photoshop. 

 

Not sure about PI.

Edited by xiando, 01 July 2017 - 05:34 PM.

[pbkoden]

In Pixinsight I've done this plenty of times. The weighting controls in the integration tool allow you to balance the different exposures. Outside of Pixinsight, I'm not sure how to do it, but it should be possible. Even if you have to go with even weighting, the extra data may help.

[SKYGZR]

Do it in DSS all the time..put (1) frame in Group1, then the rest of that same exposure time in Group 2, then others in 3,4,5 etc

 

Check the star threshold (will use the frame in G1 for that) to get around 100-150 stars, then register all first prior to stacking. This will show the one that has the best score, right click on that one & use as reference frame. Also ought to scroll down & look for the ones under a score of 100..I'll usually delete those from the system. Once the best score is selected as the reference frame, & others tossed, proceed to stack..

[xiando]

 

Do it in DSS all the time..put (1) frame in Group1, then the rest of that same exposure time in Group 2, then others in 3,4,5 etc

Do you use the longer or shorter exposure as the main group ref frame? 

[SKYGZR]

Usually will load shortest first & progress from there. Most always, the longest sub's usually contain the highest score, yet have seen that be in the intermediate exposure time(s)

Edited by SKYGZR, 01 July 2017 - 06:35 PM.

[xiando]

thanks, I'll give that a try. PS is one way to align them afterwards, but occasionally there's a tiny bit of field rotation that complicates the alignment more than I like.

[JPSTAR]

Use entropy weighting in DSS.  It's explained in the Technical Details page of DSS instructions as follows:

"Entropy Weighted Average (High Dynamic Range)
This method is based on the work of German, Jenkin and Lesperance (see Entropy-Based image merging - 2005) and is used to stack the picture while keeping for each pixel the best dynamic.
It is particularly useful when stacking pictures taken with different exposure times and ISO speeds, and it creates an averaged picture with the best possible dynamic. To put it simply it avoids burning galaxies and nebula centers.
Note: this method is very CPU and memory intensive."

[torn8o]

I did it today for the image of M20 in this month's imaging challenge (this forum). It took me a couple of tries before I got the image the way I wanted. I preprocessed everything first (ImagesPlus), got three TIF files at different exposure values and then used HDR software called Photomatix Pro. I've always used it for HDR photography and I like the way it handles ghosting and artifacts in the image. Got another TIF from that and then processed it in Photoshop for the final result. If you stack different exposure values you get more depth to the dynamic range of the image (as long as you don't whack it out ... I see so many poorly processed HDR images out there, and they have little resemblance to reality). Used well it's an exceptional tool, however.

Edited by torn8o, 02 July 2017 - 03:34 PM.

[Xshovelfighter]

Thanks all of the help! SKYGZR, how are calibration (dark, flat, bias) frames handled in the process? Is one light, dark, bias, and flat frame from one exposure put in group 1; then the rest of that exposure (lights and calibration) in group 2?

[Widotje]

Use entropy weighting in DSS.  It's explained in the Technical Details page of DSS instructions as follows:

"Entropy Weighted Average (High Dynamic Range)
This method is based on the work of German, Jenkin and Lesperance (see Entropy-Based image merging - 2005) and is used to stack the picture while keeping for each pixel the best dynamic.
It is particularly useful when stacking pictures taken with different exposure times and ISO speeds, and it creates an averaged picture with the best possible dynamic. To put it simply it avoids burning galaxies and nebula centers.
Note: this method is very CPU and memory intensive."

I'm wondering how to manage darks in this proces...I think flats can be stored in the first group, right? and do you need darks for each of the exposure times? I've experimented a little with M27. Without any darks and flats added, I get a nice picture with some amp glow on the side (i have an 178MC cool) which i can cut out of the image. When adding darks, the picture seemed to deteriorate. I only took darks that matched one of the three exposure times....anyone?

[xiando]

I wish I could help, but honestly, I've never gotten even a single good stack when using entropy-weighted. in over two dozen attempts with various image sets.

[View2]

Try APP!