Last night I was pondering how to calculate PA for any star in the image. As I said before my math abilities are very rusty by now so I thought there must some other way how to do it. Usually there is. So I come up with this.
Use centroid to find pixel coordinates of the star.
Convert FITS image to Bitmap.
Open it in some Paint program. I use Corel Paint
Calculate pixel coordinates of four corners of rectangle with the star in the centre of it.
Select and copy.
Paste as new image.
Back in Maxim convert Bitmap to FITS
Solve Plate.
And we got PA for the star of interest. The only possible error is the accuracy of pixel coordinates determined by centroid. But that error would come up anyhow when calculating PA of binary.
I have used newly created image in Reduc program to calculate PA and the separation. It comes up with believable result.
It is bit of mucking to do simple thing, but allows me to use almost any old image to measure PA and separation of the binary stars in that image. It will not work if the binary is too close to the edge of the image as there may not be enough stars in reduced image for plate solve to work.
Trigonometric solution would be easier to use as one could write small basic program to do calculations and would not have to do all that converting of the image to get same result.
|