What is the most efficient way to determine if a point is in any number of boxes

Question

I know how to check if a X,Y point is a single rectangular region, but say I have multiple regions that can potentially overlap ( the regions would have X,Y,Width,Height,Z-index ( or x1,y1,x2,y2 if that's easier -- I am not fussed on how I store that, if it's relevant )

Are there any efficient algorithms to determine if the point is inside one of the regions without having to iterate over every region.

I would prefer something without a long recalculation time when a region is added or removed, however that will be rarer then lookups.

Thank you!

Are these rectangles aligned with the x- and y-axis? – Bart Kiers Jan 20 '11 at 11:34 — Bart Kiers, Jan 20 '11 at 11:34

score 3 · Accepted Answer · answered Jan 20 '11 at 11:26

3

You could store your regions into a Quadtree (or Octree if 3D). This would help you to reject most of the regions before getting into the real collision-test.

If you have multiple layers, simply have a quadtree per layer and use the relevant one according to the layer in which your point lies.

answered Jan 20 '11 at 11:26

Vincent Mimoun-Prat

28,208
16
81
124

Thank you! That's exactly what I was looking for -- somehow I totally missed those in my own research! – Dre Jan 20 '11 at 11:32

score 1 · Answer 2 · answered Jan 20 '11 at 11:19

1

Volumetric kd-trees?

answered Jan 20 '11 at 11:19

Gareth Rees

64,967
9
133
163

What is the most efficient way to determine if a point is in any number of boxes

2 Answers2