Astrobin can plate solve my image but not PI - hints?

[IamTomzo]

Greetings

I have an image of the Rosette nebula that astrobin was able to platesolve without problems, but PI won't for some reason.   I am not getting enough star pair matches.  Clearly I am doing something wrong here.  Have any of you experienced this with the plate solve system in PI?   I was hoping to apply some photmetric color calibration to my image.

One thing I did notice is that when I acquire the focal length and pixel size from the image they appear to be incorrect.   I am imaging at 510mm (Orion 80ED with 0.85 FR) and use a Canon 500D which has a pixel size of 4.7 microns from my understanding.   PI reports the focal length at 996 mm and a pixel size of 9.  I have tried it with PI's values and values I enter directly with the same result.

I have used the tool in PI to set the coordinates on NGC2244 and those look correct.

I am sure that I must be missing something simple here - any hints?

Thanks

[pfile]

if you post a link to the xisf, someone can figure it out

rob

[IamTomzo]

Thanks Rob - I saved a file but it was about 180 MB and Astrobin could not handle it.   How do folks make a "small" xisf file out of PI?

Tom

[IamTomzo]

I think I may have figured this out - I was informed that PI only will do platesolving on linear images and mine was already stretched.   I will try it again on a non-stretched image.

Tom

[Andres.Pozo]

PixInsight can solve nonlinear images. It is a bit harder, but it does work. You can try to increase the noise reduction parameter. The documentation has also some tips to help solving difficult images.

 If you share your image with something like dropbox I can analyze it. You can also use WeTransfer and send me the link using a private message.

Andrés.

[IamTomzo]

Anders - I sent you a private message with the file link.

Thanks

Tom

[acmalko]

Greetings

I have an image of the Rosette nebula that astrobin was able to platesolve without problems, but PI won't for some reason.   I am not getting enough star pair matches.  Clearly I am doing something wrong here.  Have any of you experienced this with the plate solve system in PI?   I was hoping to apply some photmetric color calibration to my image.

One thing I did notice is that when I acquire the focal length and pixel size from the image they appear to be incorrect.   I am imaging at 510mm (Orion 80ED with 0.85 FR) and use a Canon 500D which has a pixel size of 4.7 microns from my understanding.   PI reports the focal length at 996 mm and a pixel size of 9.  I have tried it with PI's values and values I enter directly with the same result.

I have used the tool in PI to set the coordinates on NGC2244 and those look correct.

I am sure that I must be missing something simple here - any hints?

Thanks

Hi,
each time I had such a problem platesolving a picture, I used the script ImagePlateSolver and changed the algorithm alignment parameter from "polygons" to "triangles". Have a try.

Using PhotometricColorCalibration on a non-linear picture is not a good idea...

[IamTomzo]

Thanks Acmalco - I gave that a go on a linear image and the platesolve worked with a score of 80.33 after four iterations.   I then reset all the parameters in photometric color calibration to auto on everything I could and lo and behold it worked!

Thanks

Tom