diff options
| -rw-r--r-- | Wrappers/Python/ccpi/filters/regularisers.py | 5 | ||||
| -rw-r--r-- | Wrappers/Python/demos/demo_cpu_inpainters.py | 45 | ||||
| -rw-r--r-- | Wrappers/Python/src/cpu_regularisers.pyx | 31 |
3 files changed, 77 insertions, 4 deletions
diff --git a/Wrappers/Python/ccpi/filters/regularisers.py b/Wrappers/Python/ccpi/filters/regularisers.py index 8120f72..a07b39a 100644 --- a/Wrappers/Python/ccpi/filters/regularisers.py +++ b/Wrappers/Python/ccpi/filters/regularisers.py @@ -2,7 +2,7 @@ script which assigns a proper device core function based on a flag ('cpu' or 'gpu') """ -from ccpi.filters.cpu_regularisers import TV_ROF_CPU, TV_FGP_CPU, TV_SB_CPU, dTV_FGP_CPU, TNV_CPU, NDF_CPU, NDF_INPAINT_CPU +from ccpi.filters.cpu_regularisers import TV_ROF_CPU, TV_FGP_CPU, TV_SB_CPU, dTV_FGP_CPU, TNV_CPU, NDF_CPU, NDF_INPAINT_CPU, NVM_INPAINT_CPU from ccpi.filters.gpu_regularisers import TV_ROF_GPU, TV_FGP_GPU, TV_SB_GPU, dTV_FGP_GPU, NDF_GPU def ROF_TV(inputData, regularisation_parameter, iterations, @@ -114,3 +114,6 @@ def NDF_INP(inputData, maskData, regularisation_parameter, edge_parameter, itera time_marching_parameter, penalty_type): return NDF_INPAINT_CPU(inputData, maskData, regularisation_parameter, edge_parameter, iterations, time_marching_parameter, penalty_type) + +def NVM_INP(inputData, maskData, SW_increment, iterations): + return NVM_INPAINT_CPU(inputData, maskData, SW_increment, iterations) diff --git a/Wrappers/Python/demos/demo_cpu_inpainters.py b/Wrappers/Python/demos/demo_cpu_inpainters.py index b067b11..ab7ed2f 100644 --- a/Wrappers/Python/demos/demo_cpu_inpainters.py +++ b/Wrappers/Python/demos/demo_cpu_inpainters.py @@ -10,7 +10,7 @@ import numpy as np import os import timeit from scipy import io -from ccpi.filters.regularisers import NDF_INP +from ccpi.filters.regularisers import NDF_INP, NVM_INP from qualitymetrics import rmse ############################################################################### def printParametersToString(pars): @@ -146,4 +146,45 @@ a.text(0.15, 0.25, txtstr, transform=a.transAxes, fontsize=14, verticalalignment='top', bbox=props) imgplot = plt.imshow(ndf_inp_nonlinear, cmap="gray") plt.title('{}'.format('Nonlinear diffusion inpainting results')) -#%%
\ No newline at end of file +#%% +print ("%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%") +print ("Inpainting using nonlocal vertical marching") +print ("%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%") + +## plot +fig = plt.figure(4) +plt.suptitle('Performance of NVM inpainting using the CPU') +a=fig.add_subplot(1,2,1) +a.set_title('Missing data sinogram') +imgplot = plt.imshow(sino_cut,cmap="gray") + +# set parameters +pars = {'algorithm' : NVM_INP, \ + 'input' : sino_cut_new,\ + 'maskData' : mask,\ + 'SW_increment': 1,\ + 'number_of_iterations' :20 + } + +start_time = timeit.default_timer() +nvm_inp = NVM_INP(pars['input'], + pars['maskData'], + pars['SW_increment'], + pars['number_of_iterations']) + +rms = rmse(sino_full, nvm_inp) +pars['rmse'] = rms + +txtstr = printParametersToString(pars) +txtstr += "%s = %.3fs" % ('elapsed time',timeit.default_timer() - start_time) +print (txtstr) +a=fig.add_subplot(1,2,2) + +# these are matplotlib.patch.Patch properties +props = dict(boxstyle='round', facecolor='wheat', alpha=0.75) +# place a text box in upper left in axes coords +a.text(0.15, 0.25, txtstr, transform=a.transAxes, fontsize=14, + verticalalignment='top', bbox=props) +imgplot = plt.imshow(nvm_inp, cmap="gray") +plt.title('{}'.format('Nonlocal Vertical Marching inpainting results')) +#%% diff --git a/Wrappers/Python/src/cpu_regularisers.pyx b/Wrappers/Python/src/cpu_regularisers.pyx index 3625106..19dd707 100644 --- a/Wrappers/Python/src/cpu_regularisers.pyx +++ b/Wrappers/Python/src/cpu_regularisers.pyx @@ -26,7 +26,7 @@ cdef extern float TNV_CPU_main(float *Input, float *u, float lambdaPar, int maxI cdef extern float dTV_FGP_CPU_main(float *Input, float *InputRef, float *Output, float lambdaPar, int iterationsNumb, float epsil, float eta, int methodTV, int nonneg, int printM, int dimX, int dimY, int dimZ); cdef extern float Diffusion_Inpaint_CPU_main(float *Input, unsigned char *Mask, float *Output, float lambdaPar, float sigmaPar, int iterationsNumb, float tau, int penaltytype, int dimX, int dimY, int dimZ); -#cdef extern float NonlocalMarching_Inpaint_main(float *Input, unsigned char *M, float *Output, unsigned char *M_upd, int SW_increment, int iterationsNumb, int dimX, int dimY, int dimZ); +cdef extern float NonlocalMarching_Inpaint_main(float *Input, unsigned char *M, float *Output, unsigned char *M_upd, int SW_increment, int iterationsNumb, int dimX, int dimY, int dimZ); #****************************************************************# #********************** Total-variation ROF *********************# @@ -369,3 +369,32 @@ def NDF_INP_3D(np.ndarray[np.float32_t, ndim=3, mode="c"] inputData, Diffusion_Inpaint_CPU_main(&inputData[0,0,0], &maskData[0,0,0], &outputData[0,0,0], regularisation_parameter, edge_parameter, iterationsNumb, time_marching_parameter, penalty_type, dims[2], dims[1], dims[0]) return outputData +#*********************Inpainting WITH****************************# +#***************Nonlocal Vertical Marching method****************# +#****************************************************************# +def NVM_INPAINT_CPU(inputData, maskData, SW_increment, iterations): + if inputData.ndim == 2: + return NVM_INP_2D(inputData, maskData, SW_increment, iterations) + elif inputData.ndim == 3: + return + +def NVM_INP_2D(np.ndarray[np.float32_t, ndim=2, mode="c"] inputData, + np.ndarray[np.uint8_t, ndim=2, mode="c"] maskData, + int SW_increment, + int iterationsNumb): + cdef long dims[2] + dims[0] = inputData.shape[0] + dims[1] = inputData.shape[1] + + cdef np.ndarray[np.float32_t, ndim=2, mode="c"] outputData = \ + np.zeros([dims[0],dims[1]], dtype='float32') + + cdef np.ndarray[np.uint8_t, ndim=2, mode="c"] maskData_upd = \ + np.zeros([dims[0],dims[1]], dtype='uint8') + + # Run Inpaiting by Nonlocal vertical marching method for 2D data + NonlocalMarching_Inpaint_main(&inputData[0,0], &maskData[0,0], &outputData[0,0], &maskData_upd[0,0], + SW_increment, iterationsNumb, + dims[0], dims[1], 1) + + return outputData |