match_print: Matches an input and a reference shoeprint

In erichare/shoeprintr: Performs an automated statistical matching of two shoeprints

Description

Function to perform matching of footprints by matching all the candidate circles with input circles. This function performs initial matching between 3 circles on the input print and 27 candidate circles on the reference shoeprint. After the initial cliques, it select 2 best circles for each input circle and performs reinforcement matching on them. In reinforcement matching, full points on reference circle are considered instead of user defined. The function returns the final circle parameters and the statistics of the triangle formed.

Usage

match_print(print_in, print_ref, circles_input = NULL,
  circles_reference = NULL, ncross_in_bins = 30, xbins_in = 20,
  ncross_in_bin_size = 1, ncross_ref_bins = NULL, xbins_ref = 30,
  ncross_ref_bin_size = NULL, max_rotation_angle = 365, eps = 0.75,
  seed = 1, num_cores = 8, plot = FALSE, verbose = FALSE)

Arguments

print_in	The input print
print_ref	The reference print
circles_input	The input circles, specified as a matrix with columns (x, y, rad), or NULL to automatically generate
circles_reference	The reference circles, specified as a matrix with columns (x, y, rad), or NULL to automatically generate
ncross_in_bins	Number of bins in the input circle (See smart_sample)
xbins_in	Number of bins along each axis in the hexbin grid for the input circle
ncross_in_bin_size	Number of points to sample from each bin in the input circle
ncross_ref_bins	Number of bins in the reference circle
xbins_ref	Number of bins along each axis in the hexbin grid for the reference circle
ncross_ref_bin_size	Number of points to sample from each bin in the reference circle
max_rotation_angle	The maximum rotation angle, in degrees, for inclusion in the best circle matches
eps	Distance tolerance for declaring an edge match
seed	The random seed for reproducing results
num_cores	The number of processor cores for parallel processing
plot	If TRUE, produce a plot of the clique results
verbose	If TRUE, print out the timing results for each portion of the algorithm

Value

The statistics for the matching between the two circles

Examples

## Not run: 
data(input_example)
data(reference_example)

## Transform all the points to have (0,0) at lower left corner.
print_in <- leftcorner_cent(reference_example)
print_ref <- leftcorner_cent(reference_example)

## Perform Print Match
print_stats <- match_print(print_in, print_ref,
                          ncross_in_bins = 30, xbins_in = 20, ncross_in_bin_size = 1,
                          ncross_ref_bins = NULL, xbins_ref = 30, ncross_ref_bin_size = NULL,
                          eps = .75, seed = 1, num_cores = parallel::detectCores(),
                          plot = TRUE, verbose = FALSE)
print_stats

## End(Not run)

erichare/shoeprintr documentation built on May 10, 2019, 1:10 a.m.