Provided you have a planned mosaic, were the tiles make up a regular grid, you should add all of them to MosaicByCoordinates in one go. Then trim all the registered files. Tip - the registered files will be very large, so use the .xisf lossless compression when saving them. The black areas compress really well!By the way, what is the recommended method for mosaics of 2 X 3
Do you use Photometric Mosaic on 1 X3 (twice), then Mosaics by coordinates (+Trim) and then Photometric Mosaic to bring the two halves together ?
If not, you will have a tiles that need to blend on two sides at the same time.
Thanks
@Andres.Pozo Any thoughts on this? The Utilities menu is getting quite crowded, so would it be worth moving the excellent MosaicByCoordinates script to the 'Mosaic' sub menu instead?Hi John,
Thank you very much for your script. I have used it with great success with a 4 tiles mosaic in the cygnus constellation
Just a very small suggestion (although I'm not sure its in your hands): I think it would be more convenient and more logical to have the MosaicByCoordinates script in the "Mosaic" menu entry instead of the "Utilities" entry.
Regards
Nicolas
Superb result!Thank you very much for your script. I have used it with great success with a 4 tiles mosaic in the cygnus constellation:
https://storage.googleapis.com/dso-...3b7f36a7b259feba7fbb997d715ecb48-original.jpg
It took some times to apprehend the functioning of the script but the documentation you provided helped a lot!
Regards Nicolas
I have no objections. Both MosaicByCoordinates and MosaicPlanner should be in the Mosaic sub-menu.@Andres.Pozo Any thoughts on this? The Utilities menu is getting quite crowded, so would it be worth moving the excellent MosaicByCoordinates script to the 'Mosaic' sub menu instead?
Just a very small suggestion: I think it would be more convenient and more logical to have the MosaicByCoordinates script in the "Mosaic" menu entry instead of the "Utilities" entry.
@Juan Conejero Do you have any objections to this change? If you are happy for this to go ahead, I would be happy to make the trivial change and upload it to GitLab.I have no objections. Both MosaicByCoordinates and MosaicPlanner should be in the Mosaic sub-menu.
John, I'm about to run MosaicByCoordinates on all my 40 panels. You say above to use .xisf lossless compression when saving them, but I can't see an option to do this in the script. How were you intending that this would be achieved?...Provided you have a planned mosaic, were the tiles make up a regular grid, you should add all of them to MosaicByCoordinates in one go. Then trim all the registered files. Tip - the registered files will be very large, so use the .xisf lossless compression when saving them. The black areas compress really well!
I don't think MosaicByCoordinates can do this. However, when you use TrimMosaicTile and PhotometricMosaic on the 40 panels, you can then save the files with compression.John, I'm about to run MosaicByCoordinates on all my 40 panels. You say above to use .xisf lossless compression when saving them, but I can't see an option to do this in the script. How were you intending that this would be achieved?
Color mosaics of M31 are really tricky...Hi John
I finished putting my mosaics together for each of my filters, and after applying DBE I tried putting the RGB channels together for the first time. The result is disappointing - there are colour gradients everywhere.
I couldn't see these gradients in the individual filter masters, but when I did a boosted STF they did start to show up.
Do you know if there's a way to clean this up going forward, or do I need to go back to NSG and start again?
Here's the file if you want to have a look: https://drive.google.com/file/d/1VvvEbSSdXn9ygnbBtYIaMtjWm3UWOjN5/view?usp=sharing
Peter.
You were so close to solving this tricky problem! You need to deselect the 'Replace/Update Region' check box. This section is only used when you are replacing a section of a mosaic. It does not make sense to use this option with the 'Target' mode. I should probably add a warning for this combination.First of all, thank you for creating the script. It works very well for me. Now I encounter a problem and I don't quite know how to solve it after reading the help file.
Let me use the screen shot to explain. My goal is to use a wide-field reference image to correct for the gradient in individual mosaic panels. Overall it worked. However, along the edge of the mosaic where the wide-field reference does not fully cover a mosaic panel, I have problems.
In the screen shot, the upper-left window (target) is the mosaic panel to be removed for gradient. The upper-right window is the wide-field reference, registered to the target. So the target and the registered reference have identical dimensions and identical star positions. And as you can see, the reference does not completely cover the target and there is a small black stripe along the bottom edge.
The photometric mosaic script window shows my setting. This setting works well when the reference fully covers the target. Now when there is a gap in the reference, my naive imagination is that the gradient model will just be extrapolated to that uncovered area. Unfortunately, no matter the gradient model is extrapolated or not, I just can't get a final image that contains all the pixels from the target. The lower-left window is the result. It comes from the target, with brightness/color/gradient well matched to the reference, except that pixels in the bottom stripe are gone. I experimented a bit to try various settings, and still can't get a result image with black area along the bottom edge.
Hi Wei-HaoIf I deselect the Replace/Update Region box, an error message (the attached image) shows up and it does not allow me to proceed further. The message box asks me to swap the reference and the target, but this is the opposite of what I want. After experimenting a bit, I found the only way to bypass this error message and to proceed was to check the Replace/Update Region box. And the result is the black stripe at the bottom.
View attachment 12150
When using the solution I suggested above, you may get the following warning dialog for tiles entirely inside the wide field image:If I deselect the Replace/Update Region box, an error message (the attached image) shows up and it does not allow me to proceed further. The message box asks me to swap the reference and the target, but this is the opposite of what I want. After experimenting a bit, I found the only way to bypass this error message and to proceed was to check the Replace/Update Region box. And the result is the black stripe at the bottom.
Hi John and thanks for the suggestion.Color mosaics of M31 are really tricky...
The best option is to take (or borrow) a wide field image of the galaxy. This image must still be linear. Then:
View attachment 12149
- Register the wide field image to your completed mosaic.
- Split the wide field image into its R, G, B color components.
- For each of your registered stacks, use the appropriate color component of the wide field image as the reference and your stacked image as the target. Use the 'Mosaic Join' combination mode 'Target'. Instead of creating a mosaic, this will produce a copy of the target image, corrected for scale and gradient.
- Finally, mosaic all the gradient corrected stacks as before.
The wide field image does not need to be of high quality. It should still work even if it is noisy or of poor resolution. We are only using it to remove gradients. Ideally it should not have complex gradients though! If you use DBE to correct the gradients in the wide field image, do this before you register it to your mosiac to ensure that the black areas around the registered wide field image remain at zero.
Hope this helps, John Murphy