[MINC-users] Compute overlap?

[Andrew Janke]

>>   http://en.wikipedia.org/wiki/Dice%27s_coefficient
>
> Actually, it does not; the difference is that I use -add in the
> mincmath call, so $stats[1] becomes n_intersection, and $stats[0] is
> the sum of the relative complements. In other words, if the regions
> are A and B, you calculate the denominator as
>
>  n(A) + n(B)
>
> which is equal to
>
>  2 * n(A ^ B) + n(A\B) + n(B\A)

Ah, now I am with you....

>> Although I am sure there are more things to compare with than just
>> Dice's coefficient.
>
> That too - I'm always fascinated how many ways there are to turn the 4
> values of a 2x2 confusion matrix into a number :) voldiff calculates
> some but there are many more.

Aye, I got half way through re-implementing the core of voldiff in
minccmp for integer input volumes a while back but gave up with all
the seemingly multiple contradictory definitions of how things should
be done with a n x n confusion matrix across a variable number of
files!  (you'll see a commented out -kappa option in there).

All this talk of binary comparisons has got me all energised to finish
it off again...


a