MARS Update - 2023/12/06

vicent_peris

Administrator
Staff member
Hi all,

Today we completed another important milestone in this project: the definition of the processing workflow that will correct the gradients in the images. I want to show you a nice result on an image acquired at 400 mm focal length and a QHY600 camera.

Here we have the original image, a field around NGC1333:

NGC1333-1.jpg


The workflow we have designed is very reliable. In this case, the image is completely covered by dusty nebulas and there are gradients of multiple colors. If you look carefully, you'll even see a violet band on the bottom margin, coming from two different telescope pointings with two different gradients. Our workflow can correct these gradients, as shown below:

NGC1333-2.jpg


Though this technique is not designed to correct sharp gradients like the one at the bottom, it almost corrects it (only a thin violet edge is visible). This correction has been carried out with our experimental DSLR reference data.

Lastly, this technique is going to redefine the concept of black point in astrophotography. As you can see, the background in the resulting image above is brownish. This is not a lack of background neutralization, but the result of having an all-sky background reference. This will allow us to take into account the color pedestal imposed on the entire image by the local environment of the image.

Of course, we can always apply the traditional background neutralization technique that always refers to an area inside the image. This gives a more neutral result, capable of digging into the different hues of the dust inside the picture:

NGC1333-3.jpg


The same stretching and curves have been applied to the three images.


Best regards,
Vicent.
 
Hi Vicent,

thanks for the update on the progress and all the hard work, it really looks like it pays off!

Just one question: if I understood your elaborations in the past posts correctly, the process will correct multiplicative gradients using the GAIA data and additive gradients using the survey data.

Will a feature to correct multiplicative gradients be released prior to the final process, or are they dependent on each other?

CS Gerrit
 
I’m very interested in tracking your work on this Vincent. My widefield imaging rig is comprised of an ASI6200 camera and Canon EF200mm lens so is a good candidate for realising the benefit of these gradient removal processes. This is particularly so with dusty regions as you’ve demonstrated here. I have a number of project files ready to upload as soon as the MARS upload facility is ready.

CS,
Rodney

EDIT: I’ve just spotted the announcement for the upload portal. Will get on to this asap and upload my data.
 
Last edited:
Hi Vicent,

thanks for the update on the progress and all the hard work, it really looks like it pays off!

Just one question: if I understood your elaborations in the past posts correctly, the process will correct multiplicative gradients using the GAIA data and additive gradients using the survey data.

Will a feature to correct multiplicative gradients be released prior to the final process, or are they dependent on each other?

CS Gerrit

Hi Gerrit,

Excellent question! Both corrections (or modelings) are entirely independent. Actually, we are considering the possibility of releasing a new tool exclusive to multiplicative gradient correction based on Gaia spectrophotometric data. Such a tool is perfectly doable right now with Gaia/SP databases. This tool would be independent of the MARS project.
 
I’m very interested in tracking your work on this Vincent. My widefield imaging rig is comprised of an ASI6200 camera and Canon EF200mm lens so is a good candidate for realising the benefit of these gradient removal processes. This is particularly so with dusty regions as you’ve demonstrated here. I have a number of project files ready to upload as soon as the MARS upload facility is ready.

CS,
Rodney

Hi Rodney,

We have released the MARS Data Upload System today:


See my announcement on this forum board:

 
Hi Juan,

Thanks for the fast response.

At least from my perspective, it would be great to be able to do multiplicative and additive gradient correction separately. It would be interesting being able to analyze both phenomena separately in an image. Whether this is a single tool or one tool for both is less relevant.

The benefit of sperate processes would be an earlier release of the multiplicative gradient correction tool which would make some of us very happy 😊

CS Gerrit
 
Hi,

Just an additional note: multiplicative gradients need to be corrected first to properly correct additive gradients. Since we need that correction to build the all-sky reference, the multiplicative correction tool will be ready very soon.


Best regards,
Vicent.
 
Looks like a great proof of concept! I'm very curious what the end result of this is actually going to look like in PixInsight? I assume a new process that points to an online catalog? Is there any timeline or roadmap when that update might come?
 
Hi PixInsight Team,

I am impressed at how this new project is progressing so quickly, I cannot wait to see the final product when it releases eventually!

I do have a question regarding if MARS will be able to correct gradients in OSC images that are taken through filters such as the Optolong L-Ultimate or Antlia Triband RGB Ultra? Seeing that they aren't true colour images, I am interested to see how MARS will handle these types of images. I imagine that it will do an excellent job either way.

Zak :)
 
Looks like a great proof of concept! I'm very curious what the end result of this is actually going to look like in PixInsight? I assume a new process that points to an online catalog? Is there any timeline or roadmap when that update might come?
Hi. The tool will use a localizar database generated on the data we acquire and process to generate the reference map.

Right now we have 2/3 of the sky acquired with our pathfinder survey, which is based on a DSLR camera. We are using these data to develop the tools. The first version of these tools will use this database to make it available to the public as soon as possible. We'll start in some weeks to acquire data with dedicated cameras.
 
Hi PixInsight Team,

I am impressed at how this new project is progressing so quickly, I cannot wait to see the final product when it releases eventually!

I do have a question regarding if MARS will be able to correct gradients in OSC images that are taken through filters such as the Optolong L-Ultimate or Antlia Triband RGB Ultra? Seeing that they aren't true colour images, I am interested to see how MARS will handle these types of images. I imagine that it will do an excellent job either way.

Zak :)
Hi Zak,

For very narrow filters and OSC cameras the best solution will probably be to use the H-alpha and O-III data we want to provide. This will take longer since our pathfinder survey is being done with a DSLR camera. For the Triband filter, I guess the best approach will be to use the RGB reference data set.
 
Hi Zak,

For very narrow filters and OSC cameras the best solution will probably be to use the H-alpha and O-III data we want to provide. This will take longer since our pathfinder survey is being done with a DSLR camera. For the Triband filter, I guess the best approach will be to use the RGB reference data set.
Oh wow, that sounds like an amazing idea! It will be interesting to see how it plays out when the Ha and OIII data becomes available.

Thanks for your suggestions, really appreciate it and I cannot wait to see this survey project completed eventually!

Zak :)
 
Back
Top