Package Contents¶
The sinc and sinc-squared transform of specified dimension:
sinc1d.cpp
sinc2d.cpp
sinc3d.cpp
sinctransform.hpp
To compute the sinc transform directly:
directsinc.cpp
directsinc.hpp
Programs used in directsinc.cpp, as well as all example and testing code, to print arrays, generate random arrays, etc:
sincutil.cpp
sincutil.hpp
For computing Gauss-Legendre quadrature nodes and weights (see [2] in citations):
fastGL.cpp
fastGL.hpp
Simple usage examples for specified dimension:
example1d.cpp
example2d.cpp
example3d.cpp
Accuracy testing at precision range for specified dimension:
test1d.cpp
test2d.cpp
test3d.cpp
Matlab code (not C++ wrappers) for sinc transform:
sinc1d.m
sinc2d.m
sinc3d.m
sincsq1d.m
sincsq2d.m
sincsq3d.m
Matlab file containing precomputed constants for trapezoidal constants:
newconstants.mat
Matlab to compute Gauss-Legendre nodes and weights, and to generate a 3d Shepp-Logan phantom (credit to Greg von Winckel and Matthias Schabel, respectively; see licenses):
lgwt.m
phantom3d.m
As one application-based example, we also include annotated Matlab code to reconstruct an image from its nonuniformly sampled Fourier data, using the autoquad and sinc functions:
recon2d.m
recon3d.m
These use the following to compute the “optimal” quadrature weights for reconstruction, as described in [1] of citations, via the sinc-squared programs:
autoquad1d.m
autoquad2d.m
autoquad3d.m