arbitrary grid limits

[cristidonos]

Hi Denis,

Right now the grid is defined by low and high limits.
However, in many applications you need to account for direction as well (medical images), and you may end up with X axis ranging from 0->10 and Y axis ranging from 10->0. In this case lo= np.array([0,10]), and hi=np.array([10,0]). Can intergrid handle such lo/hi definitions?

Thanks

[denis-bz]

Hi Cristidonos,
is the outer rectangle axis-aligned with e.g. Y down, or also tilted / warped like

______
\    /    ?
 ---

In the latter case you can transform-irregular-quadrilateral-to-rectangle
(points.dot( trans.T ), fast),
then intergrid on the rect, then transform back.
I may have misunderstood completely, let me know
cheers
-- denis