1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
|
# CCPi-Regularisation Toolkit (CCPi-RGL)
**Iterative image reconstruction (IIR) methods normally require regularisation to stabilise the convergence and make the reconstruction problem more well-posed. CCPi-RGL software consists of 2D/3D regularisation modules for single-channel and multi-channel reconstruction problems. The regularisation modules are well-suited to use with [splitting algorithms](https://en.wikipedia.org/wiki/Augmented_Lagrangian_method#Alternating_direction_method_of_multipliers), such as ADMM and FISTA. Furthermore, the toolkit can be used independently to solve image denoising and inpaiting tasks. The core modules are written in C-OMP and CUDA languages, wrappers for Matlab and Python are provided.**
<div align="center">
<img src="docs/images/probl.png" height="225"><br>
</div>
## Prerequisites:
* [MATLAB](www.mathworks.com/products/matlab/) OR
* Python (tested ver. 3.5); Cython
* C compilers
* nvcc (CUDA SDK) compilers
## Package modules:
### Single-channel (denoising):
1. Rudin-Osher-Fatemi (ROF) Total Variation (explicit PDE minimisation scheme) **2D/3D CPU/GPU** (Ref. *1*)
2. Fast-Gradient-Projection (FGP) Total Variation **2D/3D CPU/GPU** (Ref. *2*)
3. Split-Bregman (SB) Total Variation **2D/3D CPU/GPU** (Ref. *5*)
4. Total Generalised Variation (TGV) model for higher-order regularisation **2D CPU/GPU** (Ref. *6*)
5. Linear and nonlinear diffusion (explicit PDE minimisation scheme) **2D/3D CPU/GPU** (Ref. *8*)
6. Anisotropic Fourth-Order Diffusion (explicit PDE minimisation) **2D/3D CPU/GPU** (Ref. *9*)
7. A joint ROF-LLT (Lysaker-Lundervold-Tai) model for higher-order regularisation **2D/3D CPU/GPU** (Ref. *10,11*)
### Multi-channel (denoising):
1. Fast-Gradient-Projection (FGP) Directional Total Variation **2D/3D CPU/GPU** (Ref. *3,4,2*)
2. Total Nuclear Variation (TNV) penalty **2D+channels CPU** (Ref. *7*)
### Inpainting:
1. Linear and nonlinear diffusion (explicit PDE minimisation scheme) **2D/3D CPU** (Ref. *8*)
2. Iterative nonlocal vertical marching method **2D CPU**
## Installation:
The package comes as a CMake project so you will need CMake to install it. After checking out the repository to a directory, i.e. `CCPi-Regularisation-Toolkit`, create a build directory. Additional flags to CMake will allow you to create the Matlab wrapper and/or the Python wrapper and the CUDA regularisers. In the following example I will write a line to build both wrappers and CUDA.
```bash
git clone https://github.com/vais-ral/CCPi-Regularisation-Toolkit.git
mkdir build
cd build
cmake ../CCPi-Regularisation-Toolkit -DCONDA_BUILD=OFF -DBUILD_MATLAB_WRAPPER=ON -DBUILD_PYTHON_WRAPPER=ON -DBUILD_CUDA=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=<your favourite install directory>
make install
```
The library may be used directly from C/C++ as it is compiled as a shared library. Check-out the include files in `Core`. We provide wrappers in Python and Matlab.
### Python
#### Python binaries
Python binaries are distributed via the [ccpi](https://anaconda.org/ccpi/ccpi-regulariser) conda channel. Currently we produce packages for Linux64, Python 2.7, 3.5 and 3.6, NumPy 1.12 and 1.13.
```
conda install ccpi-regulariser -c ccpi -c conda-forge
```
#### Python (conda-build)
```
export CIL_VERSION=0.10.1
conda build Wrappers/Python/conda-recipe --numpy 1.12 --python 3.5
conda install ccpi-regulariser=${CIL_VERSION} --use-local --force
cd demos/
python demo_cpu_regularisers.py # to run CPU demo
python demo_gpu_regularisers.py # to run GPU demo
```
#### Python build
If Python is not picked by CMake you can provide the additional flag to CMake `-DPYTHON_EXECUTABLE=/path/to/python/executable`.
### Matlab
If Matlab is not picked by CMake, you could add `-DMatlab_ROOT_DIR=<Matlab directory>`.
#### Linux
Because you've installed the modules in `<your favourite install directory>` you need to instruct Matlab to look in those directories:
```bash
PATH="/path/to/mex/:$PATH" LD_LIBRARY_PATH="/path/to/library:$LD_LIBRARY_PATH" matlab
```
By default `/path/to/mex` is `${CMAKE_INSTALL_PREFIX}/bin` and `/path/to/library/` is `${CMAKE_INSTALL_PREFIX}/lib`
#### Windows
On Windows the `dll` and the mex modules must reside in the same directory. It is sufficient to add the directory at the beginning of the m-file.
```matlab
addpath(/path/to/library);
```
#### Legacy Matlab installation
```
cd /Wrappers/Matlab/mex_compile
compileCPU_mex.m % to compile CPU modules
compileGPU_mex.m % to compile GPU modules (see instructions in the file)
```
### References to implemented methods:
1. [Rudin, L.I., Osher, S. and Fatemi, E., 1992. Nonlinear total variation based noise removal algorithms. Physica D: nonlinear phenomena, 60(1-4), pp.259-268.](https://www.sciencedirect.com/science/article/pii/016727899290242F)
2. [Beck, A. and Teboulle, M., 2009. Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems. IEEE Transactions on Image Processing, 18(11), pp.2419-2434.](https://doi.org/10.1109/TIP.2009.2028250)
3. [Ehrhardt, M.J. and Betcke, M.M., 2016. Multicontrast MRI reconstruction with structure-guided total variation. SIAM Journal on Imaging Sciences, 9(3), pp.1084-1106.](https://doi.org/10.1137/15M1047325)
4. [Kazantsev, D., Jørgensen, J.S., Andersen, M., Lionheart, W.R., Lee, P.D. and Withers, P.J., 2018. Joint image reconstruction method with correlative multi-channel prior for X-ray spectral computed tomography. Inverse Problems, 34(6)](https://doi.org/10.1088/1361-6420/aaba86) **Results can be reproduced using the following** [SOFTWARE](https://github.com/dkazanc/multi-channel-X-ray-CT)
5. [Goldstein, T. and Osher, S., 2009. The split Bregman method for L1-regularized problems. SIAM journal on imaging sciences, 2(2), pp.323-343.](https://doi.org/10.1137/080725891)
6. [Bredies, K., Kunisch, K. and Pock, T., 2010. Total generalized variation. SIAM Journal on Imaging Sciences, 3(3), pp.492-526.](https://doi.org/10.1137/090769521)
7. [Duran, J., Moeller, M., Sbert, C. and Cremers, D., 2016. Collaborative total variation: a general framework for vectorial TV models. SIAM Journal on Imaging Sciences, 9(1), pp.116-151.](https://doi.org/10.1137/15M102873X)
8. [Black, M.J., Sapiro, G., Marimont, D.H. and Heeger, D., 1998. Robust anisotropic diffusion. IEEE Transactions on image processing, 7(3), pp.421-432.](https://doi.org/10.1109/83.661192)
9. [Hajiaboli, M.R., 2011. An anisotropic fourth-order diffusion filter for image noise removal. International Journal of Computer Vision, 92(2), pp.177-191.](https://doi.org/10.1007/s11263-010-0330-1)
10. [Lysaker, M., Lundervold, A. and Tai, X.C., 2003. Noise removal using fourth-order partial differential equation with applications to medical magnetic resonance images in space and time. IEEE Transactions on image processing, 12(12), pp.1579-1590.](https://doi.org/10.1109/TIP.2003.819229)
11. [Kazantsev, D., Guo, E., Phillion, A.B., Withers, P.J. and Lee, P.D., 2017. Model-based iterative reconstruction using higher-order regularization of dynamic synchrotron data. Measurement Science and Technology, 28(9), p.094004.](https://doi.org/10.1088/1361-6501/aa7fa8)
### References to Software:
* If software is used, please refer to [11], however, the supporting publication is in progress.
### Applications:
* [Regularised FISTA iterative reconstruction algorithm for X-ray tomographic reconstruction with highly inaccurate measurements (MATLAB code)](https://github.com/dkazanc/FISTA-tomo)
* [Regularised ADMM iterative reconstruction algorithm for X-ray tomographic reconstruction (MATLAB code)](https://github.com/dkazanc/ADMM-tomo)
* [Joint image reconstruction method with correlative multi-channel prior for X-ray spectral computed tomography (MATLAB code)](https://github.com/dkazanc/multi-channel-X-ray-CT)
### License:
[Apache License, Version 2.0](http://www.apache.org/licenses/LICENSE-2.0)
### Acknowledgments:
CCPi-RGL software is a product of the [CCPi](https://www.ccpi.ac.uk/) group and STFC SCD software developers. Any relevant questions/comments can be e-mailed to Daniil Kazantsev at dkazanc@hotmail.com
|
