% Imported from get_default_edge_detection_settings.m
%
skipDoubleTanhAdjustment = 1;
method = 'Otsu';

[otsuApproxSettings]
globalThreshTF = true;
smoothingWindowLen = 5;
imcloseHalfGapLen = 5;
numThresholds = 1;
minNumThresholdsFgShouldPass = 1;

%  stop hardcoding these things:
%  D & E in fit for
%  f(x) = A + F *(tanh( (x - B) * D ) - tanh( (x - C) * E ))
%  are being constrained between these values
%  with initial estimate as given by
%   tanhStretchFactorInitGuess = 1.0;
%   minTanhStretchFactor = 0.5 * tanhStretchFactorInitGuess;
%   maxTanhStretchFactor = 1.5 * tanhStretchFactorInitGuess;
%  but these should probably be parameters that are calculated
%  dynamically using setting information such as the
%  point spread function's sigma width and maybe the expected
%  fluctuations in the  molecule's movement over the duration
%  of the exposure time for kymograph timeframe intensity
%  profiles or some values based on numerical analysis
%  Also Note that if the curve is rescaled away from the 
%  current resolution (pixels at some constant resolution of
%  pixel lengths in nm), these values ought to be adjusted 
%  even if there is no change in wavelength/PSF width/velocity/
%  stillness/exposure time, but currently that isn't accounted
%  for
% 
[tanhSettings]
tanhStretchFactorInitGuess = 1.0;
minTanhStretchFactor = 0.5;
maxTanhStretchFactor = 1.5;