I guess your mean a kind of error indication? In the inspector window there is a button to measure distortion. That gives the standard deviation of the star errors v.s. the solution at the bottom of the image and in the log. Something like this:
00:03:28 Pixel->Sky error inside 0.4768" or 0.1914 pixel using 115 stars. Outside 0.5290" or 0.2124 pixel using 466 stars.
00:03:28 Sky->Pixel error inside 0.4769" or 0.1914 pixel using 115 stars. Outside 0.5294" or 0.2125 pixel using 466 stars.
Default It is not measured but a separate action.
RMS is something used by PHD2 but not a fully correct term. They measure the standard deviation between the mean position and the fluctuations.
Han
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Plate-solve run at 21-May-25 10:29:55 PM
Use File/Save As to save this file
D:_images\M106NINA\RGB\green\24042025\M106_NIGHT_24042025_FILTER_green_0001.fits:
929 image stars found.
469 catalog stars found in the ATLAS / PanSTARRS (w/Gaia) catalog.
Solved using 202 of max 500, RMS residual is 0.08 arcsec, order =4
Solution took 4.8 seconds
Centerpoint RA = 12h 18m 57.213s Dec = 47° 18' 17.02"
Pointing error 0.1 arc minutes
WCS: Roll = -38.26 HScale = 0.581 VScale = 0.581
PA = 218.262°
FWHM = 1.55 arcsec
ZeroPoint = 19.82 (1 sec.)
1 solved, 0 failed.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I looked into my code but I do not see a way to extract residual error during the solving. What I showed is in the menu the backward way to test the solution. This is done by comparing the measured and documented celestial star positions using the found solution. ASTAP gives then four error values. Sky to pixel and pixel to sky for the center and out regions. I assume Pinpoint and PlateSolve3 do the same but only a generic value. Generic could be skewed for large optical distortions. But I assume you could compare them with some caution.
PlateSolve3 also reports the average residual for their 3th order solution.
Best way to compare solvers would be to compare the found star positions in ra, dec with the documented Gaia celestial position for several stars in several images from different systems.
Han
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi, I also made some python scripts using astropy in order to compare ASTAP and PinPoint to gaia 3 positions, and it seems ASTAP is better. I still have to go over the code and check things out. The test was done on ~184 stars in a singe image.
What stood out is that when I cross matched the centroids from pinpoint and astap there was a 1 pixel diferrence for every centroid.
So I need to ask if ASTAP considers the first pixel in the image to be at position (0,0) or (1,1), because I need that pixel index when coverting from pixel postion to sky position.
To quote from astropy:
Here, origin is the coordinate in the upper left corner of the image. In FITS and Fortran standards, this is 1. In Numpy and C standards this is 0.
Hello,
Can ASTAP report the RMS value after the solve?
Hi Adrian,
I guess your mean a kind of error indication? In the inspector window there is a button to measure distortion. That gives the standard deviation of the star errors v.s. the solution at the bottom of the image and in the log. Something like this:
00:03:28 Pixel->Sky error inside 0.4768" or 0.1914 pixel using 115 stars. Outside 0.5290" or 0.2124 pixel using 466 stars.
00:03:28 Sky->Pixel error inside 0.4769" or 0.1914 pixel using 115 stars. Outside 0.5294" or 0.2125 pixel using 466 stars.
Default It is not measured but a separate action.
RMS is something used by PHD2 but not a fully correct term. They measure the standard deviation between the mean position and the fluctuations.
Han
I had in mind the pinpoint report example:
Plate-solve run at 21-May-25 10:29:55 PM
Use File/Save As to save this file
D:_images\M106NINA\RGB\green\24042025\M106_NIGHT_24042025_FILTER_green_0001.fits:
929 image stars found.
469 catalog stars found in the ATLAS / PanSTARRS (w/Gaia) catalog.
Solved using 202 of max 500, RMS residual is 0.08 arcsec, order =4
Solution took 4.8 seconds
Centerpoint RA = 12h 18m 57.213s Dec = 47° 18' 17.02"
Pointing error 0.1 arc minutes
WCS: Roll = -38.26 HScale = 0.581 VScale = 0.581
PA = 218.262°
FWHM = 1.55 arcsec
ZeroPoint = 19.82 (1 sec.)
1 solved, 0 failed.
The error is default not reported.
I looked into my code but I do not see a way to extract residual error during the solving. What I showed is in the menu the backward way to test the solution. This is done by comparing the measured and documented celestial star positions using the found solution. ASTAP gives then four error values. Sky to pixel and pixel to sky for the center and out regions. I assume Pinpoint and PlateSolve3 do the same but only a generic value. Generic could be skewed for large optical distortions. But I assume you could compare them with some caution.
PlateSolve3 also reports the average residual for their 3th order solution.
Best way to compare solvers would be to compare the found star positions in ra, dec with the documented Gaia celestial position for several stars in several images from different systems.
Han
I have compared the solvers again. See this post:
https://www.cloudynights.com/topic/963963-is-astap-still-the-best-plate-solver-for-nina/page-2#entry14146651
Hi, I also made some python scripts using astropy in order to compare ASTAP and PinPoint to gaia 3 positions, and it seems ASTAP is better. I still have to go over the code and check things out. The test was done on ~184 stars in a singe image.
What stood out is that when I cross matched the centroids from pinpoint and astap there was a 1 pixel diferrence for every centroid.
So I need to ask if ASTAP considers the first pixel in the image to be at position (0,0) or (1,1), because I need that pixel index when coverting from pixel postion to sky position.
To quote from astropy: