In the beginning of 2018 I decided to do a different astrophotography project. The initial idea was a rough measurement of the proper motion of a star.

Every time I was at my observing site (and I didn't forget it...) I took 30 10s exposures of the same star. After a few series, I started to process the data.

First of all, since we are talking about small fractions of a pixel, the integration was not starightforward. It was not clear if a color image or just the best channel (green in my case), or which demosaicing method must be used, or is there any improvement if some noise reduction is applied. So the integration was done in several ways.

Also, different registering methods were tried (close stars, bright stars, default registering).

Meanwhile I derived a method, which calculates the best-fit star trajectory (using least squares, with 5 parameters: parallax, proper motion vector and an offset vector). I implemented it in Python, also with some graphical output. Out of the many differently processed data the one with the best fit was chosen: simple VNG demosaicing with Pixinsight's default registering (using only the green channel actually gives results which are closer to actual values). The measurement itself was done by using dynamic alignment in Pixinsight with the already registered pictures, selecting the star, and check it's exact coordinates (the "Edit instance source code" button will give a more precise value than what is initially showed).

So, the end result is (in parentheses are the Gaia measurements):

Fitted parameters:

------------------------------

Proper motion Dec [mas] : 10340.39 (10362.565)

Proper motion RA [mas] : -868.82 (-802.805)

Initial X position from Sun : 9.37 <- offset vector

Initial Y position from Sun : 6.40 <- offset vector

------------------------------

Parallax [mas] : 578.39 (547.4506)

Distance [parsec] : 1.72 (1.826)

Distance [lightyears] : 5.64 (5.958)

Standard deviation [pixels] : 0.029

My pixel scale is 2.151"/px, so the error in parallax is 0.014 pixel.

Obviously I didn't want to make my life harder - the star in question is the Barnard's star, which has the biggest proper motion in the sky and it is also the closest to us in the northern hemispere.