am i missing a step/configuration in preprocessing ?

[mgirdwood]

Hi
Over the last couple of days i have tried to process some tests using an Atik314L mono with LRGB filters having used OSC previously.
Only a few subs of each filter due to the lack of clear skies and it was just a test.  Using M42 due to its brightness 5 subs each at 120secs for each filter. there is data and the nebulosity is visible in all sub colours.

I have followed the tutorials available online both written and those available on Youtube. 
Created Master Flats per channel, Dark, SuperBias.
Calibrated, aligned, integrated each set of subs individually to create the LRGB files.
Using PI for each step and using the main defaults in each of the step processes
My problem is that when using the LRGB combine tool with the integrated subs i do not appear to get a combined colour image - the resultant output remains apparently Grey.   Using the histogram tool i can see that the Lines are showing that there is separate RGB.  no manipulation using the histogram or curves brings  any visible combined colour change.

I realise that this is very minimal data but with M42 there ought i believe to be something ?  Which makes me think i am missing a step in the process . ...

any pointers would be really appreciated as i would like to be able to use Mono in the future

thanks - Mark

[John_Gill]

Hi,

After Calibration, StarAlignment and ImageIntegration you end up with 4 masters of L, R, G, B.  You now need to stretch each master with ArcsinhStretch or MaskedStretch or HistogramTransformation. Then use LRGBChannelCombination to create the LRGB image.

space is not black
John

[mgirdwood]

Hi John
tried the suggestions but again i am not convinced that the integrated files are actually correct

I wonder if anyone might have a minute to review the RGB integrated ?  perhaps they will point me down the correct route .....   Subs are not so good as i took them at the end of the latest storm we just had here, so quite windy.....

https://drive.google.com/drive/folders/1hhxLIyoV9qMfj_TQRqDjc88JSwiCMt8l?usp=sharing

thanks

[John_Gill]

mmmh, I had a look at the data and I am still a bit puzzled.....anyway, try this...

For each filter set, go thru the following workflow --->
ImageCalibration
* Cosmetic Correction (not required for this test)
* Debayer (only required for Color Images)
* SubFrameSelector (not required for this test)
StarAlignment
ImageIntegration

That should generate a single image for  each RGB channel.
These 3 images should be stretched with ArcsinhStretch/MaskedStretch/HistogramTransformation
This will convert the images from linear to non-linear and then use LRGBcombination.

Put these 3 images in Google Drive and I will have another look ...
Please also post the make and model of your camera.

space is not black
John

[mgirdwood]

Hi John
thanks for looking
i have redone all the subs L R G B Hb that were taken that evening, in case the filters are all to pot.
one set simple Cal, Align, Integrate . ( percentile clipping, with Drizzle)
then the same files with a ScreenTrasnfer-Histogram stretch

scope was ED80 with x.85reducer, Atik314L mono

thanks
Mark

[WillB]

Hi Mark.

Very odd.

The FITS header for your linked google drive images states that images were taken with a ZWO183C not an ATIK mono.

Are the subs muddled somehow and you are trying to process the wrong set?
The ZWO183C images have been captured binned 2x2 so no chance to recover the colour information.

William.

[mgirdwood]

Hi Will
hadnt noticed the FITS info but on checking it seems to be the info in APT which was previously set to ZWO 183, didnt realise that would overwrite the info coming form the camera itself.

[WillB]

Hi Mark.

I had another look at just the red, green and blue singles, the pixel count, pixel size (binned) and sensor format is consistent with an Atik 314L (6.45um x6.45um 1391x1039) however loading the three frames into PI's subframe selector and looking at the image statistics it appears that the filter wheel did not move between the various filters, the statistics for the three frames are almost identical.

The FITS headers for the three subs states that they were indeed taken with red, green and blue filters selected but the statistics for the frames, and particularly with that target, shows (I think) that the filter wheel did not physically move between filter changes.

I do not know APT but possibly something was not correct in the equipment setup, or there is a problem with the filter wheel itself.

Verify that the filter wheel is working correctly and that APT is moving the correct filters into position when requested.

The only other way you could end up with almost identical red, green and blue integrated frames is if the individual frames in each integration were mixed up and you had some red frames mixed up with blue etc when you integrated the individual colour stacks of subs.

Try taking a single uncalibrated raw sub of each colour, star align and then colour combine. The resulting image will be noisy and weak but strong colour should be present. If colour is present then go back and look at the individual subs used in the integration stacks to make sure they are not muddled.

HTH

William.

[mgirdwood]

Thanks William, not sure how to check for certain if the filter wheel is turning when its attached short of taking one photo, removing the wheel and taking it apart.  but i will def see if thats an option.

[WillB]

not sure how to check for certain if the filter wheel is turning.

Hi Mark.

1. Set up a sequence of five second exposures with each LRGB filter in turn, repeat the sequence five times, shine a flash light down the OTA while looking into the mouth of the OTA, you should be able to watch the filter wheel rotate to each position as the sequence progresses, no need to remove the camera from the OTA. This will show at least that the wheel is functional and you will probably be able to confirm visually the colour of the filter by looking for a reflection from the camera sensor behind the filter.

If you can’t be sure which colour filter is in position then the following test can be used:

2. If the filter wheel appears to be rotating correctly then point the telescope in the daytime to a bright red object that is around 300mm (12”) away from the front of the OTA and larger than the aperture so that it completely fills the FOV. Take an image at short exposure time with the red filter so that the image is not saturated. If the image is saturated at minimum exposure time reduce the OTA aperture by stopping down with a cardboard or opaque plastic or cloth  (etc) diaphragm. There is no need to focus.
Once the image shows around half max pixel saturation with the red filter take an image at the same exposure time with the blue and green filters and compare the images in the statistics module. For a bright red object the average pixel value for the green and blue filters will be much smaller than the red at identical exposure times.
Repeat with a bright blue object and a bright green object, if the correct filters are rotating to the light path then the strongest average pixel value for each filter will correspond to the primary colour of the object in front of the OTA.
You can do this type of test anytime without disturbing the setup.

HTH.

William.

[mgirdwood]

Thanks William - that did the trick, the filters were working using SGP but not in APT, reset the ASCOM driver and recalibrated and its now working. 
Much Appreciated

Mark