These modifications to the parameters in the body of ss_edges.m will decrease processing time/memory requirements but may result in not finding all scale-space edges:

 

resample_factor = 4;    % Upsampling ratio.

Changing this to 2 will reduce time and memory requirements by a factor of ~4 but will decrease the edge localization accuracy and miss small scale edges unless they are well separated.
 

ks = [ .6 1.6 2.6 ];        % Initial search scale power (scale = 2^ks) list.
spacing = [ 4  4 4 ];    % Search grid spacing list, at each initial search scale.
 

Changing these to ks = [ .6 ] and spacing = [ 8 ] will reduce the time required by a large factor at the expense of missing some exclusively large scale segments and some short segments.
 

rc_Crop = [-1 0 0 0];    % [ y_min y_max x_min x_max ] where [ -1 0 0 0]
                                      % indicates complete image.

Changing this to [ 1 100 1 100 ] will only find edges in the top-left corner of the image, although preprocessing (generation of scale-space sample planes) will be performed on the whole image.

G1_init_thresh = 0.0;    % Normalized gradient threshold for starting an edge segment.
 
Changing this to G1_init_thresh = .005 will reduce the number of low salience short segments significantly, and speed up the the search a by a small factor (~10%, depending on the character of noise in the image). In principle, the edge tracker finds all edges, regardless of gradient magnitude, so this should be set to 0.0.  In practice, edges that have no points with a gradient above ~.005 have no physical significance (they are due to noise).