Open chraibi opened 5 years ago
In Gitlab by @chraibi on Dec 17, 2016, 08:19
Original comment by zhang on 15 Dec 2015, 13:46
@schroederbenjamin In JPSreport, average density and velocities are calculated over measurement area. Yes, your are true. the individual velocities are under-evaluated for stairs. I would use the velocities along the stairs to calculate the mean values. But for density calculation, I would remain use the projected area. Since the stair doesn't enlarge the effective size of the measurement area. What do you think?
In Gitlab by @chraibi on Feb 21, 2017, 14:00
We need to generalize all the methods in order to use 3D points (x, y, z)
. Actually, z
is not used, therefore the velocity calculation is not accurate.
In Gitlab by @chraibi on Dec 10, 2015, 08:39 [origin]
When using bounding boxes on stair subrooms, the projected (2D) walking speed(s) get(s) evaluated. Effectively, the walking speeds are higher as the agents move in z-direction as well.
proposted change: inclusion of an attribute "angle" that can be used to correct the results geometrically.
geometry.xml ini.xml jpsreport.xml