diff options
Diffstat (limited to 'Readme.md')
| -rw-r--r-- | Readme.md | 13 |
1 files changed, 4 insertions, 9 deletions
@@ -1,4 +1,4 @@ -# FISTA Reconstruction (Daniil Kazanteev) +# FISTA Reconstruction (Daniil Kazantsev) # General Description @@ -14,7 +14,6 @@ Software for reconstructing 2D/3D x-ray and neutron tomography datasets. The dat ### Demos: * Demo1: Synthetic phantom reconstruction with noise, stripes and zingers - * Demo2: Synthetic phantom reconstruction with noise, stripes, zingers, and the missing wedges * DemoRD1: Real data reconstruction from sino_basalt.mat (see Data) * DemoRD2: Real data reconstruction from sino3D_dendrites.mat (see Data) @@ -26,7 +25,7 @@ Software for reconstructing 2D/3D x-ray and neutron tomography datasets. The dat ### Main modules: * FISTA_REC.m – Matlab function to perform FISTA-based reconstruction - * FISTA_TV.c – C-omp function to solve for the weighted TV term using FISTA + * FGP_TV.c – C-omp function to solve for the weighted TV term using FGP * SplitBregman_TV.c – C-omp function to solve for the weighted TV term using Split-Bregman * LLT_model.c – C-omp function to solve for the weighted LLT [3] term using explicit scheme * studentst.m – Matlab function to calculate Students t penalty with 'auto-tuning' @@ -34,17 +33,13 @@ Software for reconstructing 2D/3D x-ray and neutron tomography datasets. The dat ### Supplementary: * zing_rings_add.m Matlab script to generate proj. data, add noise, zingers and stripes - * add_wedges.m script to add the missing wedge to existing sinogram - * my_red_yellowMAP.mat – nice colormap for the phantom + * my_red_yellowMAP.mat – nice colormap for the phantom * RMSE.m – Matlab function to calculate Root Mean Square Error - * subplot_tight – visualizing better subplots - * ssim_index – ssim calculation - + ### Practical advices: * Full 3D reconstruction provides much better results than 2D. In the case of ring artifacts, 3D is almost necessary * Depending on data it is better to use TV-LLT combination in order to achieve piecewise-smooth solution. The DemoRD2 shows one possible example when smoother surfaces required. * L (Lipshitz constant) if tweaked can lead to faster convergence than automatic values - * Convergence is normally much faster when using Fourier filtering before backprojection * Students’t penalty is generally quite stable in practice, however some tweaking of L might require for the real data * You can choose between SplitBregman-TV and FISTA-TV modules. The former is slower but requires less memory (for 3D volume U it can take up to 6 x U), the latter is faster but can take more memory (for 3D volume U it can take up to 11 x U). Also the SplitBregman is quite good in improving contrast. |