Horsehead processing help please

[michael8554]

Filter-removed 450D, 11x600 seconds and 21x300 seconds ISO 800 Lights, this is an 8-bit copy of the DSS autosave.tiff

I have Paint Shop Pro X3, which has Curves and Layers.

I'm finding that after stretching in Curves the dimmer stars behind the Horse are red, and if I make those grey with Curves then the background of the Horse is not very red.

Suggestions would be appreciated.

Michael

 

 

[michael8554]

After some Curves the bright stars are still white, but have pink halo's, and the fainter stars are red...

Michael

[michael8554]

Well it has occurred to me that perhaps the red background stars are red because they are shining through the red nebula.

Checked some of the on-line images and found many were like this, but some clever folk had replaced the stars with nice white round stars without flare.

So I'm going to press on with the 16bit version of above, with the view of getting more saturation, tightening the stars, and reducing the noise.

Michael

[ollypenrice]

32 minutes ago, michael8554 said:

Well it has occurred to me that perhaps the red background stars are red because they are shining through the red nebula.

Checked some of the on-line images and found many were like this, but some clever folk had replaced the stars with nice white round stars without flare.

So I'm going to press on with the 16bit version of above, with the view of getting more saturation, tightening the stars, and reducing the noise.

Michael

Whoa, how do you know what folks have done with their stars and star colour? I did nothing whatever with my star colour other than do a normal colour balance for the whole image. I don't like to post images on someone else's thread but the new forum seems to present a link as an image. Whatever, this image has had no differential star colour processing whatever. The colour balance of this image was done globally and this is the star colour (and nebular colour) that came out. 

In your image, for whatever reason, there is very little colour separation across the board and this is your problem. The first image you posted is a kind of monochrome beige. Why? I don't know. But the Flame is the same colour, in your image, as the Ha region around the Horse and this simply isn't right. The Flame is a kind of yellow-orange and the region around the Horse is classic Ha, so deep red. The nebula between the Horse and the Flame (NGC2023) is a reflection nebula and so, without any doubt at all, blue. But in your capture it is the same colour as the rest of the image. So the question is, why are you not, at present, capturing and differentiating colour? Maybe you are, and some aspect of your stacking/pre-processing/post processing is killing it but, in the images you are posing here, there is no colour information. These are monochrome images (which does not mean greyscale, let me stress.)

I think it unlikely that there really is no colour information in your capture. Our job, here, is to find out how to find it! But don't imagine that people invent it.

Olly

[michael8554]

Thanks Olly, I appreciate your opinion of my images, as I'm at a loss to understand the lack of saturation.

I recall posts describing how DSS images are desaturated and have to be improved in PS etc - but my attempts leave a lot to be desired.

I'd love to know what I'm doing wrong, as your image is very much to my liking.

What is the effect of the various Channel Background settings? I'll have to try them.

Now regarding your first comment. I seem to recall posts describing how to filter out the stars, improve them, and layer them back into the original.

Probably wrong of me to assume what some of the on-line images appear to have done to them - but are you saying nobody does that?

Michael

 

[Oddsocks]

On 14 February 2016 at 00:46, michael8554 said:

 

Filter-removed 450D, 11x600 seconds and 21x300 seconds ISO 800 Lights, this is an 8-bit copy of the DSS autosave.tiff

 

 

 

When you say "Filter removed" what filter are you talking about?

If this is the primary DSLR IR blocker have you installed the modified IR blocker in it's place?

I see no diffraction spikes in the image so I am guessing this is taken with a refractor or prime lens in which case there will be quite a bit of unfocussed IR captured by the detector which will push the histogram right over to the left and do nasty things to the colour balance....there are star haloes in the image so perhaps there is some kind of filter fitted unless they are reflections off the back of a flattener or prime lens, or just unfocussed IR haloes from the stars?

Have you carried out a new white balance since removing the filter?

It's all questions this game.....

[michael8554]

Hi Oddsocks

The IR Cut filter was removed, leaving the Low-pass filter in place.

I didn't replace the IR Cut filter with a clear glass filter or a different IR Cut filter.

Imaging with a F6.3 SCT, hence the lack of diffraction spikes.

I've read that the IR from some stars will be a problem with refractors, hadn't thought it would be a problem with a "mirror" OTA.

No, I haven't done a custom white balance as I thought that the processing would take care of that?

Today I've tried stacking the 600 sec lights, darks, flats, in about a dozen different setups in DSS.

I found that Super Pixels and Average Stacking gave best results, with signs of the blue in the small nebula that Olly drew attention to, but the Flame is still red rather than Olly's yellow.

Still work to be done on improving the noise, LX200 mount so nice tight stars will always be a challenge.

But I'm beginning to think my lights don't have enough contrast, and as Olly pointed out they are almost mono  so very little colour info to tease out.

I will also do a custom white balance and take a daylight shot of my garden to prove the camera is working correctly, yes I know that's a 1/250th sec exposure versus 10 minutes.

Michael