Closed Gabriel-p closed 7 months ago
See implementation of Voronoi algorithm here: https://github.com/Gabriel-p/voronoi
Implementation using KDE here: https://gist.github.com/Gabriel-p/17e67b705d12f7c05293
The Voronoi algorithm to detect the maximum spatial overdensity in a frame should work as follows:
params_input.dat
file the user should select between the modes low, mid, high, man
, which determine how the overdensities detection will work.auto
and low, mid, high
are selected then the algorithm should detect an overdensity if these conditions are met:
low
: all cells have at most 1.5*average_area
and a minimum of 10 members (defined as cells sharing at least one vortex)mid
: all cells having at most average_area
and a minimum of 25 membershigh
: all cells having at most 0.5*average_area
and a minimum of 100 membersman
: select both parameters p*average_area
& N_min
members(total members)/(total area)
.semi
is set, then use low, mid, high, man
the same way but use the center
& radius
values in the semi_input.dat
file to assign as the main overdensity that which lies closer to the center value within the radius given. If only the center value is given then search the entire frame. If none is given, default back to auto
global mode and select the overdensity with the highest concentration..User should handle this
Check the Voronoi tessellation method:
Somewhat related to #164.