MPO Canopus - AutoMeasuring Images
In MPO Canopus, "AutoMeasure" means matching (or aligning) a chart based on
catalog stars so that for any star on the chart and its counterpart on the image, the X/Y
distance and angle from chart center is the same as the X/Y distance and angle from image
center. In other words, the star has the same polar coordinates on the chart and image.
This is done in Canopus by using a two-star alignment method. Two stars in the image
are selected which have counterparts in the initial chart. The stars should not be too
close together in order to allow the best possible scaling. Also, a line joining the two
stars should not be vertical or horizontal, i.e., not parallel to either the X or Y axis,
otherwise it's difficult to determine the rotation of the image from true North/South.
In the manual process, you tell Canopus which two stars on the image matches two stars
on the chart. The auto-matching method determines the two stars each on the chart and
image automatically. Canopus then uses the X/Y positions of the four stars to determine
how to scale, translate, and rotate the chart so that it matches the image.
Once the chart is matched, Canopus can then match any star in the chart to a
counterpart on the image or vice versa, unless of course there is no
counterpart. This can occur for any number of reasons, e.g., the chart catalog being
incomplete to the limiting magnitude of the image or the star is a nova or asteroid. The
screen shot below shows the result of an AutoMeasure and the asteroid being automatically
measured as well (the asteroid is the one with yellow apertures about it). The cross
on the chart indicates the center of the image, not necessarily that of the target.

The numbers on the chart correspond to entries in the data table on the Reductions
page, which is shown below but for a different asteroid than in the image above.

AutoMeasure Requirements
A bonus of the AutoMeasure feature is that if the image is of an asteroid, Canopus
tries to measure the position of the asteroid automatically, saving you the time and
effort of locating the asteroid on the image and manually indicating its position. Before
Canopus can do this, it must be able to read the FITS or SBIG header "OBJECT"
keyword and interpret it correctly as the number, name, or designation of an asteroid.
Since there are so many variables, this may not always be possible. In which case, you can
tell Canopus which object it's supposed to try to find before the final steps of the
AutoMeasure routine are executed.
The pattern matching algorithm in Canopus (and MPO
Connections as well) is fairly forgiving. However, if it's not given a good first
estimate of the image scale and image orientation, the algorithm may make a match that
provides incorrect astrometry. The configuration form allows you to specify several
parameters such as focal length of the system, number and size of the pixels, and so
forth.