Difficulty with SPCC in PixInsight - Forums

Hi gang,

Unless I'm completely out of whack, there must be something in PixInsight's Spectrophotometric Color Calibration (SPCC) that's eluding me. As you can see in the image, ImageSolver has all the correct information for the image. I've downloaded the up-to-date Gaia catalogue (DR3), etc., but I still the error message that plate solving fails. I've tried all three Gaia catalogues, and get the same fail result.

Is there something obvious that I'm missing? If so, please kindly point it out before I call the Rubber Room Squad.

Many thanks,
Steve

Try with Automatic catalog selected instead. Always works for me.

Hi Andrea,
Just gave it a try - unfortunately, same result. Many thanks for the suggestion.

Steve,

I sometimes have the same issue with some pictures - they just won't solve with that PI script (although they instantly solve with ASTAP).

I noted that you do not have the box "distortion correction" checked - maybe that helps.
When I run into probelms, I also try and play around with the settings in the "Advenced Parameters" tab, sometimes that helps, often it doesn't.

It would be nice if there was an offline solver like ASTAP available in PI - this would put an end to many probelms, I guess.

As a short-time user of PI, I wonder, why this software is so highly appreciated.
Just look at the endless problems with WBPP on the PI forum…

A possible workaround would be to make PI take over the coordinates from the fits-header of your RAW file. This can be done somehow, but have not tried that yet.

Chris

Hi gang,

Unless I'm completely out of whack, there must be something in PixInsight's Spectrophotometric Color Calibration (SPCC) that's eluding me. As you can see in the image, ImageSolver has all the correct information for the image. I've downloaded the up-to-date Gaia catalogue (DR3), etc., but I still the error message that plate solving fails. I've tried all three Gaia catalogues, and get the same fail result.

Is there something obvious that I'm missing? If so, please kindly point it out before I call the Rubber Room Squad.

Many thanks,
Steve

Hello, make sure that RA/DEC Coordinates are -not 100% accuracy is needed- correct, if not overwrite the values and retry.

Also, it seems that you have not set the GAIA database XPSD files correctly, check on that also or use Automatic, as @andrea tasselli stated!

Are you sure the pixel size and focal distance are correct for your camera and telescope respectively?

I had the same problem. In Gaia script, you need to select Gaia DR3 in the Data release field even if it shows that it is allready selected.

Janez

Check the GAIA tool configuration in Pixinsight. I found that after installing the DR3/SP database for using SPCC that I had to go back and reconfigure the links for the regular DR3 database. The DR3/SP database will not work with scripts or tools that were written to work with the earlier DR3 database like Image Solver. Once I had the links to both databases configured properly with the GAIA tool everything worked as normal again. That means both databases have to be installed on your PC. Using the auto catalog as suggested before will work but it downloads from an online source each time you use it so it takes longer and will not work without internet connection. The jpg below shows how I have mine configured. Just use the wrench icon on the GAIA tool to open the tab to select links to the databases on your PC.
Gaia.jpg

Have you cropped the image? if you crop the image the image scale will not match and the solve will fail

David Cruz:
Have you cropped the image? if you crop the image the image scale will not match and the solve will fail

It won't. Cropping doesn't affect the image scale.

Hi gang,

Unless I'm completely out of whack, there must be something in PixInsight's Spectrophotometric Color Calibration (SPCC) that's eluding me. As you can see in the image, ImageSolver has all the correct information for the image. I've downloaded the up-to-date Gaia catalogue (DR3), etc., but I still the error message that plate solving fails. I've tried all three Gaia catalogues, and get the same fail result.

Is there something obvious that I'm missing? If so, please kindly point it out before I call the Rubber Room Squad.

Many thanks,
Steve

Hello Steve,
From what I’m seeing you are using the wrong Gaia database and had you marked “automatic catalog” more then likely it would have picked the right catalog for you as @andrea tasselli had mentioned, that is provided you have the correct set of Gaia databases installed.
The set you need are the Gaia DR3/SP as these are the databases that contain the SP information and thus why they are specified with the /SP on the end. Here is a screenshot from PI’s website with the database files circled in red. I personally used the “full set” but many use the smaller set with good luck.

Dale

andrea tasselli:
David Cruz:
Have you cropped the image? if you crop the image the image scale will not match and the solve will fail

It won't. Cropping doesn't affect the image scale.

It will affect, if for example you put your original focal length in a cropped image it's not going to solve.

David Cruz:
It will affect, if for example you put 2000mm focal length in a cropped image it's not going to solve.

No, if you use the right focal length, in the OP case 2800 mm.

As mentioned ImageSolver and SPCC require different GAIA catalogues. ImageSolver uses DR3 and SPCC uses DR3/SP (if you have space available download the full catalogues).

And make sure you put each catalogue in its correct place when configuring GAIA:

In GAIA DR3 > Gaia DR3
In GAIA DR3/SP > Gaia DR3/SP

Don't put anything in Gaia EDR3 (at the beginning I made the mistake of putting there the DR3 and of course, it didn't work).

And in ImageSolver make sure you select the Automatic Catalog option.

Christian Koll:
Steve,

I sometimes have the same issue with some pictures - they just won't solve with that PI script (although they instantly solve with ASTAP).

I noted that you do not have the box "distortion correction" checked - maybe that helps.
When I run into probelms, I also try and play around with the settings in the "Advenced Parameters" tab, sometimes that helps, often it doesn't.

It would be nice if there was an offline solver like ASTAP available in PI - this would put an end to many probelms, I guess.

As a short-time user of PI, I wonder, why this software is so highly appreciated.
Just look at the endless problems with WBPP on the PI forum...

A possible workaround would be to make PI take over the coordinates from the fits-header of your RAW file. This can be done somehow, but have not tried that yet.

Chris

Many thanks, Christian - yes, it would make things much simpler. So far, I have experimented with checking and unchecking the various boxes, with no luck, but I have an idea that I may not have enough catalogues, per Dale P's suggestion.

Dimitris Kavallieratos:
Hi gang,

Unless I'm completely out of whack, there must be something in PixInsight's Spectrophotometric Color Calibration (SPCC) that's eluding me. As you can see in the image, ImageSolver has all the correct information for the image. I've downloaded the up-to-date Gaia catalogue (DR3), etc., but I still the error message that plate solving fails. I've tried all three Gaia catalogues, and get the same fail result.

Is there something obvious that I'm missing? If so, please kindly point it out before I call the Rubber Room Squad.

Many thanks,
Steve

Hello, make sure that RA/DEC Coordinates are -not 100% accuracy is needed- correct, if not overwrite the values and retry.

Also, it seems that you have not set the GAIA database XPSD files correctly, check on that also or use Automatic, as @andrea tasselli stated!

Thanks, Dimitris. I've gone back and checked the correct box. RA and DEC coordinates are spot-on, per the FITS header. I'll download the full GAIA catalogues, per Dale's suggestion. Since I'm shooting at a long focal length, and there are few stars around some of them, perhaps the greater magnitude values will help.

Dale Penkala:
Hi gang,

Unless I'm completely out of whack, there must be something in PixInsight's Spectrophotometric Color Calibration (SPCC) that's eluding me. As you can see in the image, ImageSolver has all the correct information for the image. I've downloaded the up-to-date Gaia catalogue (DR3), etc., but I still the error message that plate solving fails. I've tried all three Gaia catalogues, and get the same fail result.

Is there something obvious that I'm missing? If so, please kindly point it out before I call the Rubber Room Squad.

Many thanks,
Steve

Hello Steve,
From what I’m seeing you are using the wrong Gaia database and had you marked “automatic catalog” more then likely it would have picked the right catalog for you as @andrea tasselli had mentioned, that is provided you have the correct set of Gaia databases installed.
The set you need are the Gaia DR3/SP as these are the databases that contain the SP information and thus why they are specified with the /SP on the end. Here is a screenshot from PI’s website with the database files circled in red. I personally used the “full set” but many use the smaller set with good luck.

Dale

Hi Dale, and thank you for the suggestion. In checking a Youtube video on setting up SPCC, it was suggested that only the smaller catalogue should be necessary. Because I shoot at a long FL, and since some of the galaxy targets have few stars (and they're faint), it's obvious that the full set of catalogues is necessary to increase magnitude range - the failure note I'm getting currently is that there are not enough examples for GAIA to analyze, so I'll download the full set now. My thanks, again.

andrea tasselli:
Try with Automatic catalog selected instead. Always works for me.

Thank you, Andrea. Your suggestions was certainly valuable, and I'll use the Automatic Catalogue as I experiment further.

David Cruz:
andrea tasselli:
David Cruz:
Have you cropped the image? if you crop the image the image scale will not match and the solve will fail

It won't. Cropping doesn't affect the image scale.

It will affect, if for example you put your original focal length in a cropped image it's not going to solve.

Hi David. No, I save any cropping for the 'last act' of processing, but thank you for the suggestion.

As mentioned ImageSolver and SPCC require different GAIA catalogues. ImageSolver uses DR3 and SPCC uses DR3/SP (if you have space available download the full catalogues).

And make sure you put each catalogue in its correct place when configuring GAIA:

In GAIA DR3 > Gaia DR3
In GAIA DR3/SP > Gaia DR3/SP

Don't put anything in Gaia EDR3 (at the beginning I made the mistake of putting there the DR3 and of course, it didn't work).

And in ImageSolver make sure you select the Automatic Catalog option.

Thank you, Marcelof. I'll certainly make sure of correct catalogue placement. I'm currently downloading the Full set, per Dale P's suggestion, and yes, I'll be sure to select the Automatic Catalogue.

If you drizzled the data, and then manually entered the optic information, based on your 'default' setup, the image scale will be wrong and it won't solve. You have to divide your pixel size by your drizzle iterations to get it correct. For example 3.76u becomes 1.8u for a 2x drizzle. I ran into this a couple of times when I 'reassembled' stars/starless frames and tried to Image Solve.

I also found that the image scale gets tweaked sometimes in the WBPP process, I suspect as a result of the autocrop. So if you bring in astronomic data from raw fits file, it won't align correctly with your XISF file image. However, if you get it from the WBPP master, it will be fine.

Don't know if any of these apply to you, just some learnings I had along the way.

Clayton

Christian Koll:
Steve,

As a short-time user of PI, I wonder, why this software is so highly appreciated.
Just look at the endless problems with WBPP on the PI forum...

Most times the problem lies between the computer and the back of the chair.

I've gone back and checked the correct box. RA and DEC coordinates are spot-on, per the FITS header.

I have ever seen spurious coordinates in a FITs header when an image failed to solve. Probably some misstep in setting keywords during acquisition or something..

Hi everyone,
My thanks for all your input. After making sure of everything on my end, and checking all of your suggestions, I consulted with Adam Block. Although I have managed to finally get an astrometric solution, SPCC still fails, with the constant error of "not enough samples". However, PCC (the older Photometric Color Calibration) does work, utilizing the older APASS catalogue. Adam suggested that the reason for this is because, although the databases for GAIA are large, the data does not cover many areas of the sky (in particular the one where my image is). The older APASS catalogue does cover most of the sky, and while PCC might not be as accurate as SPCC, it is what I have to work with, since I shoot with a long focal length. I have tried PCC on several images, with success every time, and I'm happy with the results.

So for anyone having difficulty with SPCC on a long focal length image, PCC is the option, unless the GAIA catalogue happens to cover the area of the sky where you're shooting,

Again, my thanks for your efforts and suggestions.
- - Steve

Hi everybody,

an interesting discussion!

In the meantime I also found a solution if Image Solver in PI would not solve your image:
You may try to solve a single exposure (subframe) instead and then transfer the coordinates to your image stack using the "Copy Coordinates"-script in PI:

Script -> Utilities -> CopyCoordinates (Source image = image with coordinates, check "Transform copy") and draw the triangel on the target image.

CS
Chris