bug in curvature

[Iago López Grobas]

Hello,

I think there is a problem in how MicrobeJ calculates the curvature.

I have a single bacterium in a timelapse. The bacterium is bent with a known curvature. When I calculate it with MicrobeJ, the result is fine if I check the result frame by frame but when I go to the data for the medial axis to check the values within each frame, I find that the values of the curvature are at least one order of magnitued off for any of the points in the medial axis. I adjusted the 'delta points' to get a result as close as possible to the real one and I did manage to do it in the value reported by MicrobeJ frame by frame but not in the values reported for the medial axis.

I hope it can be solved,

Iago.