Processing M78 Mosaic Dataset

Reggie Jones

Hi.  It's my first time attempting to process a mosaic.  Studied the documentation for Photometric Mosaic and proceeded to align and integrate via NSG all the sub exposures and eventually created 2 RGB combined images - one for the Top and one for the Bottom sub exposure data which I assume become mosaic tiles.  The next step is to use Image Solver script to plate solve the mosaic tiles.  When I did - I received the error message below.  Also - the pre requisite section said applying color calibration was not needed, but wondering if DBE should be applied to the tiles before this step and if that is the issue?

Thanks in advance

*** Error: Error: Position.Position(): invalid argument type: numeric value expected.

Please check the following items:

• The initial coordinates should be inside the image.
• The initial resolution should be within a factor of 2 from the correct value.
• Adjust the star detection sensitivity parameter, so that the script can detect most of the stars in the image without mistaking noise for stars.
• The catalog should be matched to the image. Choose the appropriate catalog and magnitude filter, so that the number of stars extracted from the catalog can be similar to the number of stars detected in the image.

*** Error: Unable to plate solve image: Alignment failed.

This usually happens because the initial parameters are too far from the actual metadata of the image.

** Warning: Process finished with errors:

M78Top_RGB.xisf: The image could not be plate solved

run --execute-mode=auto "/Applications/PixInsight/src/scripts/JohnMurphy/mosaic/src/PhotometricMosaic.js"

Processing script file: /Applications/PixInsight/src/scripts/JohnMurphy/mosaic/src/PhotometricMosaic.js

BlockHead

The Image solving problem means something is not quite right with regards to the initial coordinates or the plate scale. (Sometimes people confuse themselves because they drizzle images.)

Without having the image in hand, I cannot say specifically what the issue is. 

In the advanced parameters you should output the image to see what stars ImageSolver is seeing..this can explain a lot. You also might want to uncheck "use high quality stars"... if you data is suboptimal.

Regarding DBE..it certainly couldn't hurt to run DBE on each frame. I did I created my M78 two frame mosaic.

-the Blockhead

Reggie Jones

Ok thanks - I double checked the parameters you mentioned (I never use drizzle but I double checked that I didn't have drizzle files after alignment) and the advanced parameters.  I did change one item - I used the active window instead of adding the files.  When I ran it again, I got an error that info for the date of the image (assume from the FITS header) didn't match.  I corrected the date in the window and ran again.  Looks like it ran as advertised and solved both RGB files (Top and Bottom).  On to the next step.

* Reached maximum number of iterations without further improvement.

* Successful astrometry optimization. Score = 423.35

Saving keywords...

Saving properties...

Image Plate Solver script version 5.6.2

===============================================================================

Referentiation matrix (world[ra,dec] = matrix * image[x,y]):

-1.72325557e-04 -2.71651289e-06 +2.67472134e-01

+2.70341730e-06 -1.72308727e-04 +2.59149840e-01

WCS transformation ....... Linear

Projection ............... Gnomonic

Projection origin ........ [1528.046193 1527.959674] px -> [RA: 5 46 37.136 Dec: +0 15 59.18]

Resolution ............... 0.620 arcsec/px

Rotation ................. 0.904 deg

Reference system ......... ICRS

Observation start time ... 2000-01-01 12:00:00 UTC

Focal distance ........... 3989.53 mm

Pixel size ............... 12.00 um

Field of view ............ 31' 36.0" x 31' 36.0"

Image center ............. RA: 5 46 37.138 Dec: +0 15 59.15

Image bounds:

top-left .............. RA: 5 47 41.331 Dec: +0 31 32.09

top-right ............. RA: 5 45 34.937 Dec: +0 32 01.83

bottom-left ........... RA: 5 47 39.335 Dec: -0 00 03.55

bottom-right .......... RA: 5 45 32.947 Dec: +0 00 26.19

Reggie Jones

Ok - spoke too soon.  Ran MosaicByCoordinates and got the error below.  Apparently the Top RGB Tile is Ok but the Bottom RGB Tile has a problem; could not read WCS Coordinates.  I've sent a DropBox link with the Tiles.  Just 

Reading image:

/Users/reggiejones/Pictures/Astro Imaging/TelescopeLive/M78 CHI-1/M78 CHI1 Integrated/M78Top_RGB.xisf

Loading image: w=3056 h=3056 n=3 RGB Float32

9 image properties

ICC profile extracted: 'Display', 3284 bytes.

27 FITS keyword(s) extracted.

* Loaded astrometric solution:

Referentiation matrix (world[ra,dec] = matrix * image[x,y]):

-1.72320154e-04 -2.72190835e-06 +2.67464343e-01

+2.70267268e-06 -1.72309577e-04 +2.59153488e-01

WCS transformation ....... Linear

Projection ............... Gnomonic

Projection origin ........ [1528.000957 1527.965995] px -> [RA: 5 46 37.136 Dec: +0 15 59.17]

Resolution ............... 0.620 arcsec/px

Rotation ................. 0.903 deg

Reference system ......... ICRS

Observation start time ... 2021-01-12 12:00:00 UTC

Focal distance ........... 3989.58 mm

Pixel size ............... 12.00 um

Field of view ............ 31' 36.0" x 31' 36.0"

Image center ............. RA: 5 46 37.136 Dec: +0 15 59.15

Image bounds:

top-left .............. RA: 5 47 41.329 Dec: +0 31 32.10

top-right ............. RA: 5 45 34.939 Dec: +0 32 01.83

bottom-left ........... RA: 5 47 39.330 Dec: -0 00 03.55

bottom-right .......... RA: 5 45 32.945 Dec: +0 00 26.19

Metadata of '/Users/reggiejones/Pictures/Astro Imaging/TelescopeLive/M78 CHI-1/M78 CHI1 Integrated/M78Top_RGB.xisf' OK.

Reading image:

/Users/reggiejones/Pictures/Astro Imaging/TelescopeLive/M78 CHI-1/M78 CHI1 Integrated/M78Bot_RGB.xisf

Loading image: w=3056 h=3056 n=3 RGB Float32

1 image property

ICC profile extracted: 'Display', 3284 bytes.

Metadata of '/Users/reggiejones/Pictures/Astro Imaging/TelescopeLive/M78 CHI-1/M78 CHI1 Integrated/M78Bot_RGB.xisf': The image has no WCS coordinates

*** Error [147]: /Applications/PixInsight/include/pjsr/Sizer.jsh, line 59: uncaught exception: There where errors reading the coordinates of the images

BlockHead

The bottom image has nothing in the FITS header. 

I don't know how you lost this information. 

-the Blockhead

Reggie Jones

I was afraid of that - I saw that myself but thought I must have done something wrong.  Not sure how that happened either.  I'll need to recreate that file.  Thanks for confirming.

Reggie Jones

Ok - so here is what is happening; 

I recreated the integrated channel masters for the Bottom sub exposures using NSG.  They created as normal and each file has a FITS header.  

When I create the RGB combined file via Channel Combination using the F5 Apply button, I'd get an error; the Green Integrated master was not a color file and no combined RGB could be created.  However, when I used the Global Apply instance, it created the RGB combined file but in a different shade than the Top Combined file and with no information in the FITS Header.

Sent you the files to have a look.

Reggie Jones

One additional piece of info; I decided to work on a Telescope Live 1 Click dataset of NGC 1365.  Processed through NSG/Integration of LRGB subs and when I did a combine via LRGB combo just to crop all each master LRGB files the same, I got the same issue.  The combined file had no FITS info.  I then did an RGB combo as normal and did PCC.  PCC solved and worked as expected AND the comb file has FITS info.

Wondering if I should go back, do PCC on the Top / Bottom Tiles and then use Image Solver, MosaicByCoordinates.  I'll try that next to see what happens.

Reggie Jones

So, the good news is I retraced my steps and I think I got the process to work as advertised.  Bad news is I don't think I fully completed the process nor do I fully understand all that I did.  Sent an email to book an hour of your time via office hours to discuss.