Plate solving problem...

Questions, support and discussions
Steed
Posts: 9
Joined: 18 Oct 2019, 06:54

Plate solving problem...

Post by Steed » 25 Oct 2019, 12:22

Last night, I tried ASTAP in APT for plate solving. It worked, solved my image in a few seconds, but i found a strange problem.

The Ra and Dec were correct, but the angel was wrong.



The above are the plate-solving results of PS2,


which are consistent with the results of ASPS.



The results of ASTAP are correct about the position, but the angel is inconsistent with others.

How can I fix it?

han.k
Posts: 288
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Plate solving problem...

Post by han.k » 25 Oct 2019, 15:58

Hello Steed,

I suspect the reporting in APT is to blame. I have uploaded the image to nova.astrometry.net and loaded in ASTAP. Have a look to the north east arrow of astap.
astap2.jpg
astap2.jpg (69.58 KiB) Viewed 1058 times
Can you load the image in ASTAP viewer and solve directly? Then report the north east arrow. (In settings, your images are 2.65 degree high.)
If an ASTAP solve give the same north-east arrow, then the interface is to blame and could be an effect of the platesolve2 command line. It would be better if APT implemented the ASTAP style command line. Hopefully soon.

Clear skies, Han

Steed
Posts: 9
Joined: 18 Oct 2019, 06:54

Re: Plate solving problem...

Post by Steed » 28 Oct 2019, 03:59

Hello Han,

I solve the image in ASTAP directly, and the result is the following...



The arrow points to the southeast, just like ASTAP reported in APT. But I also notice that the image in ASTAP viewer is somehow upside down. Maybe that is the problem?

Clear skies, Steed

Steed
Posts: 9
Joined: 18 Oct 2019, 06:54

Re: Plate solving problem...

Post by Steed » 28 Oct 2019, 11:13

Hello Han,

I uploaded the image to Astrometry.net, you can check the results on the following link:
http://nova.astrometry.net/user_images/ ... #annotated

han.k
Posts: 288
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Plate solving problem...

Post by han.k » 30 Oct 2019, 11:58

The solving angle in ASTAP is correct. You can swap the image (view menu ASTAP) and the angle is show identical as with the other solver. It must be the interface or jpeg/FITS difference. I will investigate further.

Later:
nova.astrometry.net is reporting 133 degrees E of N.
Astap is reporting -229 degrees equals 131 degrees E of N
PS2 is reporting 49.2 degree.
ASPS is reporting 49.2 degree (this should be identical to nova.astrometry.net ???)

The difference lies in the display of the FITS file. The origin of FITS pixel position 1,1 is left bottom pixel for most programs. APT display this pixel at left top. So the image is vertical flipped making 131 degrees equals 180-131 is 49 degrees.

What I could do is mimic the PlateSolve2 behavior.

Han

han.k
Posts: 288
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Plate solving problem...

Post by han.k » 30 Oct 2019, 20:45

I have overhauled the ASTAP angle reporting in PlateSolve2 mode. It should now fully mimic PlateSolve2 behavior taking into account things like orientation, flipped, FITS or JPEG.

Can you test it? Version 0.9.291
Clear skies, Han

Steed
Posts: 9
Joined: 18 Oct 2019, 06:54

Re: Plate solving problem...

Post by Steed » 01 Nov 2019, 06:52

Hello Han,

I tried the version 0.9.291, it seemed work right now.
WX20191101-134942@2x.jpg
thank you.

han.k
Posts: 288
Joined: 23 Nov 2017, 15:04
Location: The Netherlands
Contact:

Re: Plate solving problem...

Post by han.k » 01 Nov 2019, 11:14

Good. I will remove the images to make some space. My server space is limited. :)

Clear skies, Han

Post Reply