Detect corners with FAST

<A>: Get the centre of mass by getting average <x> and <y> value from the dataset.

<B>: For each value in the dataset, multiply <x> by 2 and <y> by 2, and take the average for that.

You should now have two points: <A> and <B>, and also the dataset.

Go through each point in the dataset and calculate the angles and distances between <A>, <B>, and <a point in the dataset>

Then run assignment problem using the angles and distances.

This should allow you to geometrically match two images.