APP 1.055 strange color issue??

Hi Betelgeuze,

Thank you for sharing this 😉

Indeed, the differences that you see between 1.054 and 1.055 are in the star color calibration module. That has been upgraded. Some remarks first:

1. I am still working on this module so it's not working yet like I think it should.
2. The new version reduces the green cast seen which was apparent in 1.054, but it still only calibrates using 3 parameters, I am working to increase this to 9 parameters and possibly even giving the user control over atmospheric and galactic extinction coefficients.
3. since we are talking about colors, it's good to know if we are seeing the same colors on our monitors.I work with 2 calibrated monitors, I calibrated them with a Spider 5 elite calibration sensor. Do you have your monitor calibrated in some way or another? If not, its well worth it to have a look at that and what options are available for monitor calibration. Most monitors come out of the factory uncalibrated, so with bad colors. Only very expensive monitors are calibrated when you buy them, and even then, you have no gaurantee that the colors are like they should. A calibration sensor gives you that assurance.

My initial response to the images is that I think that the second image probably is better calibrated than the first.

The first shows a greenish cast over the blue nebula, but I agree that the second image might be a little bit too purple. So maybe the best calibration would be somewhere between these 2?

But.... I did a quick search for a good quality image of the Iris nebula and I think that this does confirm that the second image, version 1.055, is much better calibrated.

Image found in APOD image discussions at Starship Asterisk, and technical details about the Iris Nebula. There actually is broadband red emission in the nebula causing some parts of the blue nebulosity to become purple and your image does show it at the right spots. Still maybe a bit too purple, but I think much better than your blue/greenish Iris from version 1.054.

http://asterisk.apod.com/viewtopic.php?t=33573#p228130

Iris Nebula (LBN 487 / VDB 139) with associated open cluster NGC 7023
by Steve Cannistra

<<Although not officially designated as an emission nebula, closer examination ... will reveal a linear ridge on either side of SAO 19158 that represents HII emission. The peripheral regions are comprised of reddish dust that obscures light from background stars.

In the center of the nebula, there are several ruddy-colored wisps and filaments of dust that emit broad band red light, instead of reflecting the more typical blue light of a reflection nebula. These red regions represent extended red emission (ERE), which is a type of phospholuminescence associated with dust particles that are bombared by high energy UV radiation from SAO 19158.>>

and

Thanks, Art. I'm well aware of those reddish filaments of the center of the nebula. They show up very well in this fine picture of NGC 7023 by Adam Block. But I wasn't referring to those faint filaments of emission nebulosity near the star itself, SAO 19158 (which I prefer to call HD 200775). I meant those dark reddish-brown nebulas far away from the blue reflection nebula. There are several of them in the upper right part of today's APOD, and some can be found below the Iris nebula, at about 8 and 9 o'clock. They are indeed murkily reddish in color. Does their color have anything to do with photoluminescence due to ultraviolet light? It would be the first thing I have heard about such a phenomenon.

This is the Adam Block image, although at a much higher resolution since he uses a very big telescope under excellent atmospheric conditions, the colors do match very nicely  I think.

Link from Adam's website:

http://www.caelumobservatory.com/gallery/n7023.shtml

Let's directly compare your images with this image, I rescaled and rotated your images to better match the Adam Block image (did not change saturation or stretch). Seeing this, I do think the 1.055 version is much better, both the nebula color and the star colors are better, only complaint is that it might be a little bit too red. The stars directly surrounding the Nebula are reddened by dust, which is clear as well:

This does point out, that perhaps most Iris nebula images that can be found on the web are shown with a too low red multiplication factor., causing the nebula to only appear blue... instead of blue with some purple spots.

If someone has better information on how the Iris Nebula should look like in RGB broadband then please share this.

Betelgeuze, I would be interested to receive your original integration file, so I can add it to my test images for star color calibration. Can you send it to me using wetransfer.com using support@astropixelprocessor.com ?

Kind regards,

Mabula

Posted by: Betelgeuze

Hello Mabula,

 

...

 

Also a bit unclear to me, in the new calibrate function, I can load the bias, flat and dark images, and click both “create bad pixel map” and “ create Masterbias, -dark, -flat, giving me the best master frames, or is it still better to create them manually and having more control following the DSLR work flow video, I got used to.

 

Clear and healthy sky' s!

Hi Betelgeuze,

The only thing that has changed, is that you can now have APP make the regular masters and BPM simultaneously, or choose to make only the masters or the BPM. I haven't changed internally yet, that APP figures out the best way of calibrating the data. So I would suggest to stick to your workflow that you have right now 😉

I intend to do a big upgrade in the calibration engine in the next version though, then things will change and I will provide extensive information on the changes. My goal is to make the calibration engine much smarter so it ensures good calibration no matter what you provide APP to work with.

Kind regards,

Mabula

Thank you very much for your reply, it is very interesting, and I am surprised I captured colors, I did not even know they exist..

Looking at that Iris nebula from Adam Block, I want almost sell all my imaging gear. 😉

I will take another look at my monitor calibration, you have a point there.

In the meant time.. I have send you all the needed files from my Iris nebula.

(You will see clearly my struggling with the reducer/flattener spacing.)

Thank you for investigating them.

Kind regards!

You're welcom Betelgeuze 😉

That image from Adam Block is amazing indeed...

Thanks for sending the data !

Mabula

Hi Mabula,

If I may ask,  is this "Star color calibration" issue gonna be fixed in the next 1.056 release ?

Or for the time being is it better in my case to downgrade back to 1.054 ?

Kind regards

Hi Betelgeuze,

I am almost ready to release 1.056 and I will try to include a further update of the Star Color Calibration module.

I think in 1.054 the results were too green and in 1.056 in some cases, the results were too purple. So I will at least adjust it in such a way that it is between those 2 versions.

But, like I indicated before, a bigger improvement is in the pipeline... in which the calibration is over more parameters and will be done one the selected starfield(s) whose star colors should then match the colors, which you would expect from a large population of stars of mainly main sequence stars.

Kind regards,

Mabula