GRD2DReconstruction.h File Reference

declares the stir::GRD2DReconstruction class More...

#include "stir/recon_buildblock/AnalyticReconstruction.h"
#include "stir/RegisteredParsingObject.h"
#include <string>
#include "stir/shared_ptr.h"

Go to the source code of this file.

Classes
class	stir::DiscretisedDensity< num_dimensions, elemT >
	This abstract class is the basis for all image representations. More...

Namespaces
	stir
	Namespace for the STIR library (and some/most of its applications)

Detailed Description

declares the stir::GRD2DReconstruction class

GRD2D maps each PET view into Fourier space, interpolates non-uniform samples to a Cartesian grid using a Kaiser–Bessel kernel, then applies an inverse 2D FFT and resamples to the output image. A radial low-pass noise filter can suppress high-frequency noise.

The algorithm, its reference, and comments on its implementation are described in Chapter 5 of Dimitra Kyriakopoulou's doctoral thesis, “Analytical and Numerical Aspects of Tomography”, University College London (UCL), 2024, supervised by Professor Athanassios S. Fokas (Cambridge) and Professor Kris Thielemans (UCL). Available at: https://discovery.ucl.ac.uk/id/eprint/10202525/

Author

Dimitra Kyriakopoulou