One more thing. I want to control the density of the inhibitory centers.
I think I can just lay out a specific number of inhibitory centers along the mid line of the interveinal region. The number of centers should scale with the area of that region.
Changing that value will automatically adjust the size of the cells outlined by the cross veins.
First I want to see what the current algorithm is producing for these values:
19 / 31869.399982241066 = 0.0005961831729052815
19 / 27087.796469773042 = 0.0007014228721484185
18 / 23686.508640337794 = 0.000759926263229282
16 / 20969.555177596296 = 0.0007630109396452182
16 / 18630.3655106109 = 0.0008588129948866125
13 / 15830.371058298359 = 0.0008212062719266037
11 / 12832.183990489386 = 0.0008572196290321807
12 / 10556.268198690257 = 0.0011367653581867948
8 / 9002.222885188105 = 0.000888669398884011
8 / 7965.070475583991 = 0.0010043853377723502
7 / 7644.009717750923 = 0.0009157497515661965
23 / 31869.399982241066 = 0.0007216954198327091
22 / 27087.796469773042 = 0.0008121738519613267
23 / 23686.508640337794 = 0.0009710168919040825
11 / 20969.555177596296 = 0.0005245700210060875
17 / 18630.3655106109 = 0.0009124888070670258
10 / 15830.371058298359 = 0.0006316971322512337
12 / 12832.183990489386 = 0.0009351486862169244
11 / 10556.268198690257 = 0.0010420349116712284
5 / 9002.222885188105 = 0.0005554183743025069
8 / 7965.070475583991 = 0.0010043853377723502
9 / 7644.009717750923 = 0.001177392537727967
Average is 0.000845
- Chart the distribution.
I’ve got a bit of a mess here since I’m using Numpy and Shape so I have to translate between different formats. Right now my translation code is all over the place which is not ideal. I’ll put it on the to-do list.
For now, here’s the output from the dart-throwing algorithm:
And here’s the output from the …. whatever I’m calling the algorithm that neatly arranges a bunch of points along the midline.
Looks neat as-is but what does it look like after Lloyd’s relaxation?
Oh, and I found out why length was giving me the exact same number for every primary vein. I was using Segment.length which isn’t the length of the vein but the length of the segment itself. It’s been a while since I looked at the Segment class, I forgot it actually just has a position, a direction, and a length.
So I converted the root segment and all its generations of children into a LineString and then I was able to get the length of the primary vein itself.
Phew.