Stretch gets lost in 1.5.369
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Stretch gets lost in 1.5.369
Open stacked DSLR image from DSS
Perform Autodev - stretched image is visible, looks good in monochrome.
Perform Crop
Perform Bin
Perform Wipe w/Keep - stretched image is visible, color (w/button) looks good.
Exit Wipe and return to main screen - stretched image is no longer visible, and the display looks pre-stretch.
Save the image at that point and bring it into Gimp, and there is no stretch apparent, no color.
This is quite repeatable. If you need files, let me know.
- Bob
-----------------------------------------------------------
StarTools 1.5.369
Fri Oct 25 20:51:18 2019
-----------------------------------------------------------
File loaded [C:\Processing\2019_09_Heart_Combo_CLSCCD_b\2019_0927_Heart_CLSCCD\Master_Stack_16.TIF].
Image size is 6016 x 4016
---
Type of Data: Linear and was Bayered, but not whitebalanced
--- Auto Develop
Parameter [Ignore Fine Detail <] set to [Off]
Parameter [Outside RoI Influence] set to [50 %]
Parameter [RoI X1] set to [0 pixels]
Parameter [RoI Y1] set to [0 pixels]
Parameter [RoI X2] set to [6016 pixels (-0)]
Parameter [RoI Y2] set to [4016 pixels (-0)]
Parameter [Detector Gamma] set to [1.00]
Parameter [Shadow Linearity] set to [50 %]
--- Crop
Parameter [X1] set to [44 pixels]
Parameter [Y1] set to [65 pixels]
Parameter [X2] set to [5938 pixels (-78)]
Parameter [Y2] set to [3976 pixels (-40)]
Image size is 5894 x 3911
--- Bin
Parameter [Scale] set to [(scale/noise reduction 50.00%)/(400.00%)/(+2.00 bits)]
Image size is 2947 x 1955
--- Wipe
Parameter [Mode] set to [Correct Color & Brightness]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Filter] set to [1 pixels]
Parameter [Drop Off Point] set to [100 %]
Parameter [Corner Aggressiveness] set to [100 %]
Parameter [Aggressiveness] set to [75 %]
File saved [C:\Processing\2019_09_Heart_Combo_CLSCCD_b\2019_0927_Heart_CLSCCD\foo.TIF].
Perform Autodev - stretched image is visible, looks good in monochrome.
Perform Crop
Perform Bin
Perform Wipe w/Keep - stretched image is visible, color (w/button) looks good.
Exit Wipe and return to main screen - stretched image is no longer visible, and the display looks pre-stretch.
Save the image at that point and bring it into Gimp, and there is no stretch apparent, no color.
This is quite repeatable. If you need files, let me know.
- Bob
-----------------------------------------------------------
StarTools 1.5.369
Fri Oct 25 20:51:18 2019
-----------------------------------------------------------
File loaded [C:\Processing\2019_09_Heart_Combo_CLSCCD_b\2019_0927_Heart_CLSCCD\Master_Stack_16.TIF].
Image size is 6016 x 4016
---
Type of Data: Linear and was Bayered, but not whitebalanced
--- Auto Develop
Parameter [Ignore Fine Detail <] set to [Off]
Parameter [Outside RoI Influence] set to [50 %]
Parameter [RoI X1] set to [0 pixels]
Parameter [RoI Y1] set to [0 pixels]
Parameter [RoI X2] set to [6016 pixels (-0)]
Parameter [RoI Y2] set to [4016 pixels (-0)]
Parameter [Detector Gamma] set to [1.00]
Parameter [Shadow Linearity] set to [50 %]
--- Crop
Parameter [X1] set to [44 pixels]
Parameter [Y1] set to [65 pixels]
Parameter [X2] set to [5938 pixels (-78)]
Parameter [Y2] set to [3976 pixels (-40)]
Image size is 5894 x 3911
--- Bin
Parameter [Scale] set to [(scale/noise reduction 50.00%)/(400.00%)/(+2.00 bits)]
Image size is 2947 x 1955
--- Wipe
Parameter [Mode] set to [Correct Color & Brightness]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Filter] set to [1 pixels]
Parameter [Drop Off Point] set to [100 %]
Parameter [Corner Aggressiveness] set to [100 %]
Parameter [Aggressiveness] set to [75 %]
File saved [C:\Processing\2019_09_Heart_Combo_CLSCCD_b\2019_0927_Heart_CLSCCD\foo.TIF].
Re: Stretch gets lost in 1.5.369
Hi
I think StarTools is working as expected.
All you would need to do is add Auto Dev to the end of your steps before you save if you want the StarTools stretch that you saw in Wipe retained.
Does that work for you.
I think StarTools is working as expected.
All you would need to do is add Auto Dev to the end of your steps before you save if you want the StarTools stretch that you saw in Wipe retained.
Does that work for you.
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
Hi!
Nope, not expected. I do not recall ST previously losing its stretch after Wipe, so why now? Bug.
And...
In Wipe, the [color] button shows very nice results.
Then exit, and the nebulosity is gone and we're back to a few stars-on-black.
So...redo Autodev, and carry on with the rest of the workflow...(no color evident due to luminosity, as expected)
Get to the Color module, and the colors are a mess - nothing like what was visible in Wipe's [color].
But that seems to be a problem best dealt with after we figure out why the stretch is being discarded on exit from Wipe.
- Bob
Nope, not expected. I do not recall ST previously losing its stretch after Wipe, so why now? Bug.
And...
In Wipe, the [color] button shows very nice results.
Then exit, and the nebulosity is gone and we're back to a few stars-on-black.
So...redo Autodev, and carry on with the rest of the workflow...(no color evident due to luminosity, as expected)
Get to the Color module, and the colors are a mess - nothing like what was visible in Wipe's [color].
But that seems to be a problem best dealt with after we figure out why the stretch is being discarded on exit from Wipe.
- Bob
Re: Stretch gets lost in 1.5.369
Hi Bob,
Happy-kat is right (thank you for helping Bob out ) - this is indeed expected behaviour.
The stretch you are seeing in Wipe is a temporary AutoDev (with parameters chosen to bring out as many flaws as possible) of the linear unwiped data, e.g. it is not the stretch that was previously active. "Automatic temporary AutoDev in effect" is visible under the image to highlight this.
After performing a Wipe, any previous stretch you may have had, is completely invalidated and "useless". This is because dynamic range that was previously used to encode gradients and signal bias (usually the bulk of the dynamic range) has now been freed up for the purpose of encoding/showing actual real detail. Re-doing your stretch is virtually a necessity after using Wipe. E.g. the previous stretch was made (either by AutoDev or by you) for a wildly different dataset.
In every workflow and/or StarTools tutorial from 1.3 to 1.6, re-doing your stretch has always been the first thing to do after running Wipe.
As of 1.5, StarTools reverts back to the linear (but wiped) state of the data to force you to re-do your stretch. In 1.3 and 1.4, Wipe would still try to re-apply the "useless" previous stretch to the best of its abilities.
Perhaps you can attach an image showing the "mess" and tell us the behaviour you are seeing vs what you would expect?
Thank you!
Happy-kat is right (thank you for helping Bob out ) - this is indeed expected behaviour.
The stretch you are seeing in Wipe is a temporary AutoDev (with parameters chosen to bring out as many flaws as possible) of the linear unwiped data, e.g. it is not the stretch that was previously active. "Automatic temporary AutoDev in effect" is visible under the image to highlight this.
After performing a Wipe, any previous stretch you may have had, is completely invalidated and "useless". This is because dynamic range that was previously used to encode gradients and signal bias (usually the bulk of the dynamic range) has now been freed up for the purpose of encoding/showing actual real detail. Re-doing your stretch is virtually a necessity after using Wipe. E.g. the previous stretch was made (either by AutoDev or by you) for a wildly different dataset.
In every workflow and/or StarTools tutorial from 1.3 to 1.6, re-doing your stretch has always been the first thing to do after running Wipe.
As of 1.5, StarTools reverts back to the linear (but wiped) state of the data to force you to re-do your stretch. In 1.3 and 1.4, Wipe would still try to re-apply the "useless" previous stretch to the best of its abilities.
The Colors you are seeing in Wipe are uncalibrated/unbalanced, so this tends to be the other way around...the colors are a mess
Perhaps you can attach an image showing the "mess" and tell us the behaviour you are seeing vs what you would expect?
Thank you!
Ivo Jager
StarTools creator and astronomy enthusiast
StarTools creator and astronomy enthusiast
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
Ivo,In every workflow and/or StarTools tutorial from 1.3 to 1.6, re-doing your stretch has always been the first thing to do after running Wipe.
Well, if it's WAD, it's WAD... But I would not have reported it as a bug if I expected that behavior. Missed the wipe do-over convention completely. I had been getting results with the prior behavior, so no reason to expect this change.
I'll put some color examples together.
- Bob
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
What follows is a sequence of screen shots from an abbreviated workflow. I've skipped sharpen, contrast, life, etc., just focusing on Autodev, Wipe and Color.
AutoDev result, pre-wipe
AutoDev result, pre-wipe
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
We are now in the Wipe module, and this is the monochrome result from Temporary Autodev. Looks good.
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
Still in the Wipe module - this is what the [color] button shows - a very nice starting point.
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
Exit Wipe, back at the main console...
-
- Posts: 60
- Joined: Mon Sep 24, 2018 12:30 am
Re: Stretch gets lost in 1.5.369
Here's the result of the AutoDev re-do