No colour after a simple processing

[fmrjt]

hello. I'm a new user of PI. My name is Frederic.
Three weeks ago I have made seven photos of Orion's nebula from Tarragona, an urban location. With a Nikon D5300 in raw format and exposure time less than a minute. No bias, no darks, no flats. Only lights.
I used star alignment and image integration. But I have lost colour.
After I used STF and Histogram Transformation but colour doesn't appear. Images are black and white. 
What's wrong with my processing?
Can I get colour photos after processing only with light photos?
Thank you very much.
Fred.

[pfile]

Debayer needs to be in there somewhere... did you do that?

rob

[fmrjt]

Rob, thank you for your answer. I don't apply debayer in my first work with PI. I don't know in which step I have to use, and if I have to apply over the individual photos or over the result of the integration  ?

Now, in a second attempt to improve my photos I am trying to apply it and :

If I used a integrated image (*.xisf) the error message is      Missing or invalid CFA pattern description property    when I use Apply F5

If I used a integrated image (*.xisf) the error message is    No image could be demosaiced      when I use Apply global F6

Do I have to use Apply or Apply global and when   ?

Thank you very much.

[dld]

Hello Fred and welcome,

you should use the Debayer process before the StarAlignment process. Since you don't calibrate your lights, you have to debayer, register, and then integrate.

Since sometime in the future you will want to calibrate your lights, keep in mind that calibration masters (master flats, darks, bias) and lights calibration has to be performed with raw (not debayered) data. Then, just before StarAlignment, you'll have to debayer your calibrated lights.

[fmrjt]

Hello dld for your explanations. Now I have colour. Not much colour but affortunately it isn't a black and white photo

I tell you what I have done:

VIEW   >  EXPLORER WINDOWS  >   FORMAT EXPLORER   Then I press Raw, > EDIT PREFERENCES   and I select pure raw    >  OK

PROCESS  >  DEBAYER >  ADD FILES   Then I use the default parameters I select the output directory     >   APPLY GLOBAL

PROCESS   >   STAR ALIGNEMENT   >  I try the reference image   >   ADD FILES    I use the default parameters    >  APPLY GLOBAL

PROCESS   >    IMAGE INTEGRATION     >  ADD FILES   I use the default parameters     >  APPLY GLOBAL

Now the ending image is very poor; I have read in Internet this simple procedure (and I have applied):

STF and I press the "nuclear" button.   I press the triangle icon of STF and drag to the bottom bar of Histogram Transformation

Then I drag the "New Instance" icon of HT onto my image. I turn off the "monitor" icon of STF. I have got a green image, but very interesting.

Three o four years ago someone in my astronomical association (Associació Astronòmica de Torroja del Priorat) explained to use SCNR  and I have did it.

The photo doesn't have red colour, but a little of pale blue or grey.

Thank you very much.

Anything you comment I'll try to apply.

Fred.

[dld]

Hello Fred,

I am glad you are seeing some color now. Usually the integrated images aren't that impressive regarding colors, for several reasons. One is light pollution and one other is the green dominance because of the way DSLR cameras record color. We have to follow some more steps in order to bring the proper colors back. I'll suggest viewing Astro Dude's tutorials on youtube, since he's far better than me on explaining things (and has a very calm voice ). You may start from Part 5 Dynamic Crop and ABE of his series. There are of course many other kind people who provide good advice, like Kayron Mercieca, with his tutorial on reducing light pollution.

[fmrjt]

I'll try to watch the video or web pages
Thank you very much dld. PI is a very complete program, but without help is hard to do well.

[fmrjt]

I can do a lot of steps with PIXI, but are they right?? Can I improve anything??? This is the global process for Orion's Nebula:

VIEW   >  EXPLORER WINDOWS  >   FORMAT EXPLORER   Then I press Raw, > EDIT PREFERENCES   and I select pure raw    >  OK

PROCESS  >  DEBAYER >  ADD FILES   Then I use the default parameters I select the output directory     >   APPLY GLOBAL

PROCESS   >   STAR ALIGNEMENT   >  I try the reference image   >   ADD FILES    I use the default parameters    >  APPLY GLOBAL

PROCESS   >    IMAGE INTEGRATION     >  ADD FILES   I use the default parameters     >  APPLY GLOBAL

I go to STF and I press the nuclear button

Dynamic crop and I cut the image

Automatic Background Extractor  Using division

Another time STF and I press the nuclear button

Background neutralization with a small image preview without stars

Photometric color calibration using force plate solving. I get the data of right ascension and declination for m42 using search coordinates.

Focal length for my Nikon D5300 300 mm x 1,5 = 450

Pixel size 3,89 nm

Then I have to use another time STF and the nuclear button because if i don't do it I can't use save as using the jpg format

And finally ACDNR for the noise reduction.

Some times I don't know if I have to use STF and the nuclear button. !!!!!!!!!!!!!!

Thank you very much for help me.


 

[ngc1535]

Then I have to use another time STF and the nuclear button because if i don't do it I can't use save as using the jpg format

This is an odd statement. The automatic screen stretch (the "nuclear button") adjusts the brightness levels for display on the screen. This is for your eyes only and it does not change the values of the underlying data. You certainly *can save* an unstretched image. However the resulting image will look *black* because JPEGs are 8bit images. The current STF (screen stretch) is used to create the image. So by pressing the auto STF (nuclear button), you are choosing a much better way to save the image.

Now... PI will tell you all kinds of information about how JPEGs are terrible with a loss of information etc etc... this doesn't have anything to do with the screen stretch (it will say these things every time).

-adam

[fmrjt]

Thank you very much Adam. I'm trying to do different processes, invert the order in some cases, and do new processes. Some times I can save without problems. But in some cases I have to press the nuclear button and drag the blue triangle button onto the HT, and apply to my photo....
Thank you very much.
Fred.

[STEVE333]

Hi Fred - One suggestion for when you are using STF. In the upper Left corner of the STF process there is a "chain" icon for Linking/Unlinking the RGB channels. If you make sure that icon is NOT selected, the STF will try to do color-balancing when stretching the displayed image. This is usually a much more accurate indication of the colors in your final processed image.

Steve

[fmrjt]

Steve, I'll try to do in my new processes. I have made a little and fast prove and it seems to be very interesting.
Thank you very very very  much.
Fred.

[fmrjt]

Hello Steve. I have proved to use the nuclear button without mark the "chain". The great difference is that you have the colour immediately after the first use of the nuclear button.
But when I have finished all the process the differences between images without the "chain" marked and the images using the "chain" activated are small.
It is easier to use the "chain" not marked because you can see the colours at the first steps, not at the ending steps.
Also I think that I lose some red color with the third or fourth nuclear button use. Is it posible? 
Thank you very much.

[STEVE333]

1) Hello Steve. I have proved to use the nuclear button without mark the "chain". The great difference is that you have the colour immediately after the first use of the nuclear button.

2) But when I have finished all the process the differences between images without the "chain" marked and the images using the "chain" activated are small.

3) It is easier to use the "chain" not marked because you can see the colours at the first steps, not at the ending steps.

4) Also I think that I lose some red color with the third or fourth nuclear button use. Is it posible? 
Thank you very much.

Hi Fred -

1) Yes, this is the advantage of using the STF unlinked (Chain not marked).

2) This is normal.
a) Using the STF with the linked option (Chain marked) means that the R/G/B channels will ALL be stretched the same amount, and, the same offset is applied to all three channels. Thus, if the image is not color corrected the STF displayed image will also not be color corrected.
b) Using the STF unlinked (Chain not marked) will apply different offset and stretch to each channel to try to achieve color balance.
c) However, once you have done color correction and BackgroundNeutralization in your processing, STF stretching this processed image with or without the Chain marked will show little difference because the proper offsets and gains have already been applied to the data with the processing.

3) I agree. I always use the STF unlinked.

4) Using the STF nuclear button does NOT change your data. The STF process only changes how the data is displayed on your computer, but, it does not change the data. So, no matter how many times you press the nuclear button it won't change the amount of red (or any other color) in the data itself. The data is only changed when  you apply HistogramTransformation (or one of the other stretches (like ArcsinhStretch or ExponentialTransformation, etc.).

Hope this makes sense.

Steve

[fmrjt]

Steve Thank you very much for the information. I'm proving with the STF, nuclear button unchained, with ACDNR; Photometric color Calibration; MultiscaleMedianTransform; Histogram Transformation, and everything I could use and understand only a little.
Thanks.