I've thought about the Voronoi approach but thought, in the absence of much skill at computing, it would be easier to look an insect in the eye, as it were. But it might be interesting, once I have a suitable eye and have mapped it, to see if the Voronoi approach gives the same answer. I can't see (sorry . . . ) why the insect eye should have hexagons, other than that represents the minimum amount of joints per unit area, and thus leaves the maximum area available for light to enter. Does that seem reasonable?
Julian
On 20 Sep 2012, at 16:05, Mark Jessell wrote:
> HI
>
> do they have to be perfect hexagons (which by definition I don't think thay can be anyway)? Why not take the inverse approach and seed your curved surface with points on a triangular lattice and calculate the Voronoi cells, I think (of course without trying) that you would get a nice slightly irregular but useful tiling?
>
> cheers
>
> Mark Jessell
> IRD Toulouse
|