Band Module .. where in workflow to use

Questions and answers about processing in StarTools and how to accomplish certain tasks.
OttoMann
Posts: 15
Joined: Sat Aug 03, 2013 8:46 pm

Band Module .. where in workflow to use

Post by OttoMann »

Hi all
I don't have a banding issue but a friend I'm trying to help learn ST has a pretty bad horizontal banding issue with his T2i.
I never even noticed the Band Mod. until this morning and was reprocessing my friends image to show him my workflow
and to try Band for the first time.

My plan was to do my normal flow and do Band just before color and de-noise. Once I got to it I read in the help it's best to use band
before stretching.. whoops..

So my question is when to use it?
My normal workflow is..

Bin > AutoDev > Crop > Wipe > Develop > Wipe > Contrast > Decon > HDR > Sharp > (normally I do Color here but did Band) > Color > De-Noise

So where in my workflow should I tell my friend to do Band? After Bin? After the first wipe?

Any help appreciated..
Thanks
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Band Module .. where in workflow to use

Post by admin »

OttoMann wrote:Hi all
I don't have a banding issue but a friend I'm trying to help learn ST has a pretty bad horizontal banding issue with his T2i.
I never even noticed the Band Mod. until this morning and was reprocessing my friends image to show him my workflow
and to try Band for the first time.

My plan was to do my normal flow and do Band just before color and de-noise. Once I got to it I read in the help it's best to use band
before stretching.. whoops..

So my question is when to use it?
My normal workflow is..

Bin > AutoDev > Crop > Wipe > Develop > Wipe > Contrast > Decon > HDR > Sharp > (normally I do Color here but did Band) > Color > De-Noise

So where in my workflow should I tell my friend to do Band? After Bin? After the first wipe?

Any help appreciated..
Thanks
Ideally, it would be applied it when the data is still linear and while Tracking is still off; the earlier you catch the banding issue, the fewer modules will have to work with the 'tainted' data. I still need to make the Band module a 'Tracking friendly', so that it matters less when you apply it (like most other modules when Tracking is on).

If I may ask, why do you apply Wipe twice?
Ivo Jager
StarTools creator and astronomy enthusiast
OttoMann
Posts: 15
Joined: Sat Aug 03, 2013 8:46 pm

Re: Band Module .. where in workflow to use

Post by OttoMann »

When I tried the Band module and did Help it said it's works best before the image is stretched..
I guess I assumed that meant before any stretching.. I suppose my data was still linear since I had not turned off tracking yet..
But Band didn't work very well at all in removing the banding..

Again it's not my issue just trying to help a friend..
He's on the fence about processing software and I'm trying to show him what ST can do..

He claims he gets better results getting rid of banding using PS after the fact.. trying to prove him wrong..

As far as 2 wipes.. When I was learning ST a friend at AF shared his workflow and it seemed to help to wipe twice.. the second time setting the drop off point to 0
I dunno..

This is the workflow I generally use.. At least the steps..
If you find flaws please let me know..

StarTools basic workflow

------------------------------------------------------------------------

1.Open - has to be a .fts file straight from DSS

**Type of data?** > Always click 2nd from left - Linear,was Bayered, is not whitebalanced

**Possible stacking artifacts detected** > Click ok

--------------------------------------------------------------------------------------------

2.Bin - The default is 50%, what I always use > Click Keep

---------------------------------------------------------------------------------------------

3.AutoDev - it may look weird but just go with it. We truly stretch later
Parameter [Ignore Fine Detail <] set to [Off]
Parameter [Outside ROI Influence] set to [50 %]

Click Keep

------------------------------------------------------------------------------------------------

4.Crop - you want to crop any edges that look different. Each image will be different as to
how much you have to crop. I'm using your M92. Even is you don't see any artifacts you want to
take at least a sliver off all 4 sides. So for just this Image my parameters are..

**New Image size will be 2556x1693**
X1 = 30 pixels Y1 = 54 pixels
X2 = 2586 pixels (-25) Y2 = 1747 pixels (-30)

Click Keep

------------------------------------------------------------------------------------------------

5.Wipe - **Tracking Active** > Click Ok

**Preview Out of date. Please click 'do' button to re-render image.**

Parameter [Mode] set to [Correct Color & Brightness]
Parameter [Temporary AutoDev] set to [No]
Parameter [Aggressiveness] set to [75 %]
Parameter [Drop Off Point] set to [100 %]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Filter] set to [1 pixels]
Parameter [Corner Aggressiveness] set to [100 %]

Click Do

Click Keep

-------------------------------------------------------------------------------------------------

6.Develop - **Previously Stretched** > Click Redo global stretch

Each picture will vary a bit with Digital Development but the other settings will stay the same.
I start off taking Digital Development to 91.5% .. I've gone as high as 92.5 and as low as 89.
There is no right or wrong..for this image. Still not sure about the Dark Anomaly Filter so I turn off

Parameter [Digital Development] set to [91.50 %]
Parameter [Gamma] set to [1.00]
Parameter [Dark Anomaly Filter] set to [Off]
Parameter [Dark Anomaly Headroom] set to [5 %]
Parameter [White Calibration] set to [Use Stars]
Parameter [Red Luminance Contrib.] set to [100 %]
Parameter [Green Luminance Contrib.] set to [100 %]
Parameter [Blue Luminance Contrib.] set to [100 %]
Parameter [Skyglow] set to [0 %]

Click Keep

-----------------------------------------------------------------------------------------------------

7.Wipe - **Tracking Active** > Click Ok

**Preview Out of date. Please click 'do' button to re-render image.**

I always wipe twice. The second time I change the drop off point to 0.

Parameter [Mode] set to [Correct Color & Brightness]
Parameter [Temporary AutoDev] set to [No]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Filter] set to [1 pixels]
Parameter [Drop Off Point] set to [0 %]
Parameter [Corner Aggressiveness] set to [100 %]
Parameter [Aggressiveness] set to [75 %]

Click Do

Click Keep

-------------------------------------------------------------------------------------------------------

8.Contrast - **Preview Out of date. Please click 'do' button to re-render image.**

Parameter [Compensate Gamma] set to [No]
Parameter [Expose Dark Areas] set to [No]
Parameter [Precision] set to [256 x 256 pixels]
Parameter [Dark Anomaly Headroom] set to [15 %]
Parameter [Dark Anomaly Filter] set to [1 pixels]
Parameter [Aggressiveness] set to [75 %]

Click Do

Click Keep

---------------------------------------------------------------------------------------------------------

9.Decon - **Star Mask** > Click Generate mask automatically

Parameter [Radius] set to [1.5 pixels]
Parameter [Regularization] set to [1.00 (optimal noise and detail)]
Parameter [Iterations] set to [6]
Parameter [Image Type] set to [Deep Space]
Parameter [Mask Behavior] set to [De-ring Mask Gaps, Hide Result]
Parameter [Mask Fuzz] set to [8.0 pixels]

Click Keep

-----------------------------------------------------------------------------------------------------------

10.HDR

Parameter [Algorithm] set to [Reveal DSO Core] .. one click left
Parameter [Dark/Bright Response] set to [1.00]
Parameter [Noise Suppression] set to [Off]
Parameter [Small Detail Precision] set to [Max]
Parameter [Detail Size Range] set to [50 pixels] .. this setting will vary depending on the image. Higher # lessens the reveal
Parameter [Channels] set to [Brightness Only]

Click Keep

---------------------------------------------------------------------------------------------------------------

11.Sharp - **Please click the Next button to commence scale extraction**

Structure Size = Large

Click Next

Parameter [Scale 1] set to [100%]
Parameter [Scale 2] set to [100%]
Parameter [Scale 3] set to [100%]
Parameter [Scale 4] set to [100%]
Parameter [Scale 5] set to [100%]
Parameter [Amount] set to [100%]
Parameter [Small Detail Bias] set to [75 %]
Parameter [Intelligent Enhance] set to [Yes]
Parameter [Mask Fuzz] set to [8.0 pixels]

Click Keep

--------------------------------------------------------------------------------------------------------------

NORMALLY I'D GO TO THE COLOR MODULE NOW BUT SINCE YOU HAVE THE BANDING ISSUE I'M GUESSING THIS
IS THE TIME TO TRY AND FIX IT. NEVER TRIED THIS BEFORE SO BEAR WITH ME..

--------------------------------------------------------------------------------------------------------------

12.Band - *Preview Out of date. Please click 'Do' button to re-render image.

Parameter [Orientation] set to [Horizontal]
Parameter [Algorithm] set to [Algorithm 1]

Click Do

Then I changed Parameter [Orientation] set to [Vertical]

Click Do

Click Keep

(Ok. Just FYI. I just found out you want to do Band before you stretch your image. So I should have done it
after the first Wipe I'm guessing? So step 6 should have been band? I'll have to research more.

---------------------------------------------------------------------------------------------------------------

13.Color - **Partial Mask in effect** > Click Fill Mask

This will also differ with each image. This is also what I have the most trouble with.
For this image my settings are.. Again there is no right or wrong.

Parameter [Style] set to [Scientific (Color Consistancy)]
Parameter [Saturation Amount] set to [300 %]
Parameter [Bright Saturation] set to [Full]
Parameter [Dark Saturation] set to [1.50]
Parameter [Cap Green] set to [To Yellow]
Parameter [LRGB Method Emulation] set to [Straight CIELab Luminance Retention]
Parameter [Bias Slider Mode] set to [Sliders Reduce Color Bias]
Parameter [Red Bias Increase] set to [2.51]
Parameter [Green Bias Increase] set to [1.40]
Parameter [Blue Bias Increase] set to [1.35]
Parameter [Mask Fuzz] set to [1.0 pixels]

Click Keep

---------------------------------------------------------------------------------------------------------------

Color is the last step before noise reduction.

Click the Track Button, 5 Buttons right of the open button at top of screen.

**Stop Tracking and Noise Reduction** > Click button to far left.. Stop Tracking, do final noise reduction

**Please click 'Next' button to commence scale extraction.

Parameter [Filter Type] set to [Distance Weighted Outlier Rejection]

Click Next

I pretty much use these settings all the time.

14.Wavelet De-Noise

Parameter [Smoothness] set to [82 %]
Parameter [Scale 1] set to [90 %]
Parameter [Scale 2] set to [90 %]
Parameter [Scale 3] set to [90 %]
Parameter [Scale 4] set to [90 %]
Parameter [Scale 5] set to [65 %]
Parameter [Redistribution Kernel] set to [4.5 pixels]
Parameter [Read Noise Compensation] set to [Off]
Parameter [Brightness Detail Loss] set to [12 %]
Parameter [Color Detail Loss] set to [12 %]
Parameter [Scale Correlation] set to [3]
Parameter [Mask Fuzz] set to [1.0 pixels]

Click Keep

You're done. Normally you'd click save now. Name your file. It by default saves as a .tiff
OttoMann
Posts: 15
Joined: Sat Aug 03, 2013 8:46 pm

Re: Band Module .. where in workflow to use

Post by OttoMann »

Again this is my friends pic..

this is before I tried the Band module..
https://www.dropbox.com/s/e2zgc61gy1ebu ... Tools.tiff

this one I tried the Band module but I must have done something wrong
https://www.dropbox.com/s/ph93jixmphgbv ... essed.tiff
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Band Module .. where in workflow to use

Post by admin »

Not a bad workflow at all.
OttoMann wrote:When I tried the Band module and did Help it said it's works best before the image is stretched..
I guess I assumed that meant before any stretching.. I suppose my data was still linear since I had not turned off tracking yet..
But Band didn't work very well at all in removing the banding..
Try using it on the linear image without tracking on (e.g. 'lie' to StarTools and say that it's modified). Do the Banding reduction, then turn Tracking on and process as normal.
He claims he gets better results getting rid of banding using PS after the fact.. trying to prove him wrong..
You will :) Taking care of banding before you process is the way to go. Bias frames (e.g. addressing the cause of the banding) would be even better...
As far as 2 wipes.. When I was learning ST a friend at AF shared his workflow and it seemed to help to wipe twice.. the second time setting the drop off point to 0
I dunno..
That effectively runs the Vignetting preset - this is useful if you don't have a perfectly flat field. You run the risk of removing nebulosity towards the edges though (as the aggressiveness gets progressively higher towards the edges).
Still not sure about the Dark Anomaly Filter so I turn off
The Dark Anomaly filter in the Develop module helps the module filter out any 'darker-than-real-interstellar-background' noise/pixels/anomalies. If they're present, then filtering them out, will help Develop lock on the correct black point better. Else, Develop will adopt the cautious stance that such dark pixels must be preserved and are 'real' background.
14.Wavelet De-Noise

Parameter [Read Noise Compensation] set to [Off]
Try bumping this up to smoothen your background a bit better if the noise floor is visible (it more than likely will be if you have only slightly noisy data and Develop used a conservative blackpoint).
Ivo Jager
StarTools creator and astronomy enthusiast
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Band Module .. where in workflow to use

Post by admin »

OttoMann wrote:Again this is my friends pic..

this is before I tried the Band module..
https://www.dropbox.com/s/e2zgc61gy1ebu ... Tools.tiff

this one I tried the Band module but I must have done something wrong
https://www.dropbox.com/s/ph93jixmphgbv ... essed.tiff
I tried and the Band module definitely makes a difference when applied after the fact, but I'm keen to see if it makes an even bigger difference when applied to linear data...
Ivo Jager
StarTools creator and astronomy enthusiast
OttoMann
Posts: 15
Joined: Sat Aug 03, 2013 8:46 pm

Re: Band Module .. where in workflow to use

Post by OttoMann »

Hi Ivo..
Thanks for the tips in my workflow.. appreciate it!
Still having an issue trying to figure out / get Band to remove the banding in my friends images.
He sent me a different autosave with really bad banding. He says he re-took 50 bias frames and re-stacked
the image.. think it's M5.. but the banding is still awful!. So I processed using my normal workflow 4 different ways.
1. Just my workflow.. no Band module.
2. my workflow using Band after final noise reduction.
3. I opened the image as Modified , not linear. Did Band, then turned on tracking and did my normal workflow.
4. Same as 3 but used Band after final noise reduction also.. so I ran it twice..

1.
https://www.dropbox.com/s/zocx6qa4fp4fw ... Tools.tiff

2.
https://www.dropbox.com/s/0yrsid7kvib4d ... after.tiff

3.
https://www.dropbox.com/s/hh36qghof2dz2 ... efore.tiff

4.
https://www.dropbox.com/s/k8dati98v8v03 ... after.tiff

I don't know.. I kinda think #2 looks the best.. removed the most of it.. Any ideas on what I'm doing wrong,
what I can do different..?

Thanks,
Kurt
User avatar
admin
Site Admin
Posts: 3382
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Band Module .. where in workflow to use

Post by admin »

OttoMann wrote:Hi Ivo..
Thanks for the tips in my workflow.. appreciate it!
Still having an issue trying to figure out / get Band to remove the banding in my friends images.
He sent me a different autosave with really bad banding. He says he re-took 50 bias frames and re-stacked
the image.. think it's M5.. but the banding is still awful!. So I processed using my normal workflow 4 different ways.
1. Just my workflow.. no Band module.
2. my workflow using Band after final noise reduction.
3. I opened the image as Modified , not linear. Did Band, then turned on tracking and did my normal workflow.
4. Same as 3 but used Band after final noise reduction also.. so I ran it twice..

1.
https://www.dropbox.com/s/zocx6qa4fp4fw ... Tools.tiff

2.
https://www.dropbox.com/s/0yrsid7kvib4d ... after.tiff

3.
https://www.dropbox.com/s/hh36qghof2dz2 ... efore.tiff

4.
https://www.dropbox.com/s/k8dati98v8v03 ... after.tiff

I don't know.. I kinda think #2 looks the best.. removed the most of it.. Any ideas on what I'm doing wrong,
what I can do different..?

Thanks,
Kurt
Yep, my vote would be for #2 as well.
Nothing wrong with this perse! Some small things I'd do different for this image, but that's getting into the realm of personal taste. I would probably not stretch the background so much (e.g, use AutoDev with an ROI over the glob , combined with tweaking the 'Ignore Fine Detail <' parameter so AutoDev won't optimize the stretch for the fine noise. You could also use the Isolate preset in the Life module to Isolate the glob and /or make it pop.
Any chance I/we could get my/our hands on the data and have a play?
Ivo Jager
StarTools creator and astronomy enthusiast
OttoMann
Posts: 15
Joined: Sat Aug 03, 2013 8:46 pm

Re: Band Module .. where in workflow to use

Post by OttoMann »

Sure thing..
Curious to see what you come up with..

https://www.dropbox.com/s/sshf33633njye ... estack.fts
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Band Module .. where in workflow to use

Post by MadMaxwellSmart »

Hello Ivo, I want to thank you and Kurt for helping me out with my Banding issue, I am using the trial version of Star Tools right now but I am sold on it, I have enjoyed using it...it makes me want to get back out and get some new images so I can play with it. I would like to see what you've come up with on the image Kurt sent you....Thanks so much...Glenn :thumbsup:
Post Reply