No color after using BatchPreprocessing

[Thierry Van Driessche]

I have been using PI for some weeks treating pictures taken in RAW format by using a DSLR camera.

First I used following method: image integration of bias and darks, image calibration of flats and afterwards image integration of them and finally image calibration, registration and integration of lights. After these steps I processed the image. When looking to the calibrated lights and to the masterlight, both type of images have a decent color saturation.

I shortly am trying to use a different method by using the batch preprocessing.
In the tab of the lights, “Calibrate only” is checked, all other parameters have been reset to the factory-default values.
When looking to the calibrated lights after using the processing, these are grayed out while when looking to the calibrated and registered subs using the first mentioned method, these ones are in color.

Is this normal or am I missing something or is there a setting I need to adjust?

[tgervais]

I have the same problem. If processing is done using Star Alignment instead of Batch Processing - there is no problem with color.
 
I can't help with your problem because I have the same problem.  I was hoping someone had a solution to the problem.  Like a different setting in Batch Processing.

Fingers are crossed here that someone will come along with a solution.. 

[pfile]

i think in another thread someone mentioned you forgot to debayer your images... which you have to do after calibration but before star alignment.

in BPP you want the setting "CFA Images" to be checked.

rob

[Thierry Van Driessche]

Hi Rob,

I have been using debayering already. But when looking to a debayered image, the colors are way too low. Even using curves transformation and pushing the color saturation quite high doesn’t bring the colors back like it should.

Any other idea?

[pfile]

sorry i think i was thinking of steve's thread... anyway what is the total integration time, what is the target and were you using any LP filters? what is the subexposure length and what is the optics?

thanks

rob

[oldwexi]

Hi Thierry!
Your lights    HAVE TO BE Gray     after Calibration!!!!
Yes, this is normal and how it should be.

After the Calibration you run the Debayer process on the calibrated gray images,
so they become for the first time at the right time RGB images.
After Debayer you StarAlign the debayered RGB images,
After StarAlign  you run    ImageIntegration  to the staraligned RGB images.
Then there should be enough color there for processing.

If you do wrongly the debayer job AFTER StarAlign, means to late,  you will endup with either minimal color
or a funny colormix.

Did you follow the above mentioned steps ?
When yes,  let us have one of your calibrated and debayered xisf file.
And also the integrated xisf file.   (No jpegs)

Gerald

[Thierry Van Driessche]

sorry i think i was thinking of steve's thread... anyway what is the total integration time, what is the target and were you using any LP filters? what is the subexposure length and what is the optics?

thanks

rob

Hi Rob,

In total 12 subs of 30 seconds, no filter, Canon EOS 5D Mark III with zoom 75-300 set at 300. Target is Andromeda nebulae.

Thanks, Thierry

[pfile]

sounds to me like there's very little signal in your subs - at 300mm it's probably like f/5.6 right? 30s is way too short.

rob

[Thierry Van Driessche]

Hi Thierry!
Your lights    HAVE TO BE Gray     after Calibration!!!!
Yes, this is normal and how it should be.

After the Calibration you run the Debayer process on the calibrated gray images,
so they become for the first time at the right time RGB images.
After Debayer you StarAlign the debayered RGB images,
After StarAlign  you run    ImageIntegration  to the staraligned RGB images.
Then there should be enough color there for processing.

If you do wrongly the debayer job AFTER StarAlign, means to late,  you will endup with either minimal color
or a funny colormix.

Did you follow the above mentioned steps ?
When yes,  let us have one of your calibrated and debayered xisf file.
And also the integrated xisf file.   (No jpegs)

Gerald

Hi Gerald,

Thanks for the answer. I will give it a trial. I guess I can uncheck the "CFA image" in the global options of the lights in the BPP.

Thanks and I will keep posting, Thierry

[Thierry Van Driessche]

Hi Gerald,

I followed your recommendation. It seems to work. Thanks a lot !!

BR, Thierry

[Thierry Van Driessche]

Hi Thierry!
Your lights    HAVE TO BE Gray     after Calibration!!!!
Yes, this is normal and how it should be.

After the Calibration you run the Debayer process on the calibrated gray images,
so they become for the first time at the right time RGB images.
After Debayer you StarAlign the debayered RGB images,
After StarAlign  you run    ImageIntegration  to the staraligned RGB images.
Then there should be enough color there for processing.

If you do wrongly the debayer job AFTER StarAlign, means to late,  you will endup with either minimal color
or a funny colormix.

Did you follow the above mentioned steps ?
When yes,  let us have one of your calibrated and debayered xisf file.
And also the integrated xisf file.   (No jpegs)

Gerald

Hi Gerald,

It looks like I still have trouble with the colors.

Here is a picture processed with PI using another method:
https://drive.google.com/open?id=0BwfK_I4Wluk2U3RCTThOUm9tOEk

I first integrated bias and darks, then calibrated flats and integrated them and afterwards calibration of lights, registering them and integrate them. After this, processed the image.

As you asked, here are the files you need.

One calibrated file is here
https://drive.google.com/open?id=0BwfK_I4Wluk2OWljWDE3V1Nmd0U

One debayered file is here:
https://drive.google.com/open?id=0BwfK_I4Wluk2d0Q2LW0yVXNkU3M

And the integrated file is here:
https://drive.google.com/open?id=0BwfK_I4Wluk2M3ZQeXN5VVJmMWs

Thanks for your time and BR, Thierry

[msmythers]

Thierry

I find it much easier to troubleshoot issues if you go back to the basics. The basics in this case is to stack your images manually in the simplest way. That would be just you light subs, no calibration files. So with OSC(one shot color) camera the basic flow is debayer your light subs(use batch debayer), align(regisiter) your debayered subs with StarAlignment. Integrate(stack) your debayered-registered subs with ImageIntegration. After ImageIntegration your left with an image that might not look right. In my case very green when STF is applied with it linked. Un-linked it looks fine. After BackgroundNeutralization and ColorCalibration applying STF-linked looks correct. I have not done any DynamicBackgroundExtraction(DBE) or ABE, this is just to proof color. I then adjusted color using the color saturation tool. It has a range selection which your can think of as amplification. Here are 2 images showing the steps and files-folders. This is only 5 light subs out of the 180 taken. These images were collected with a Nex-5 and are 30 second exposures. I used a 45mm camera lens at f4.5.

Things to note, my bayer pattern might be different from yours. I used default setting for StarAlignment and ImageIntegration I used Percentile clipping rejection because I only had 5 subs total. The settings used were just for proving I had color not for a quality final image.


Mike

[Thierry Van Driessche]

Thierry

I find it much easier to troubleshoot issues if you go back to the basics. The basics in this case is to stack your images manually in the simplest way. That would be just you light subs, no calibration files. So with OSC(one shot color) camera the basic flow is debayer your light subs(use batch debayer), align(regisiter) your debayered subs with StarAlignment. Integrate(stack) your debayered-registered subs with ImageIntegration. After ImageIntegration your left with an image that might not look right. In my case very green when STF is applied with it linked. Un-linked it looks fine. After BackgroundNeutralization and ColorCalibration applying STF-linked looks correct. I have not done any DynamicBackgroundExtraction(DBE) or ABE, this is just to proof color. I then adjusted color using the color saturation tool. It has a range selection which your can think of as amplification. Here are 2 images showing the steps and files-folders. This is only 5 light subs out of the 180 taken. These images were collected with a Nex-5 and are 30 second exposures. I used a 45mm camera lens at f4.5.

Things to note, my bayer pattern might be different from yours. I used default setting for StarAlignment and ImageIntegration I used Percentile clipping rejection because I only had 5 subs total. The settings used were just for proving I had color not for a quality final image.


Mike

Hi Mike,

Thanks for the reply and advice. I think I will go back indeed to basics like I was doing earlier and forget BPP.

BR, Thierry

[msmythers]

Thierry

Once you are comfortable with what the process is I see no reason not to use BPP especially for calibration.  I use BPP for the first integration just to see what my image really looks like. Then I use the BPP registered files in the ImageIntegration tool and adjust parameters for the best final image. This is fast since the original BPP Imageintegration files are cached.

The main thing is to get comfortable with the tools and workflow.



Mike