Input Sanity Check¶

The first step to estimating the blur model involves computation of the transmission function, which is the ideal radiograph image that is formed in the absence of X-ray source and detector blurs. This computation of transmission function is performed internally in the function pysaber.estimate_blur(), which is used to estimate the blur model by computing the parameters of X-ray source and detector blurs. However, this computation of transmission function is not fail-proof and may result in inaccurate edge localization if certain assumptions made when computing the transmission function are not satisfied.
Before using pysaber.estimate_blur() to estimate blur model, it is recommended to check for accurate edge localization in the transmission function. The transmission function can be computed using the function pysaber.get_trans_masks().
The function pysaber.get_trans_masks() also returns the mask arrays for the transmission function and radiograph, which are used to include or exclude certain pixels from blur estimation. By default, the radiograph mask only excludes a small number of pixels along the boundary of the radiograph from blur estimation. Additional pixels can be excluded from blur estimation by appropriately setting the input arguments of the functions pysaber.get_trans_masks() and pysaber.estimate_blur(). The mask for transmission function should also exclude the padded pixels in addition to those pixels excluded by the radiograph mask. Hence, pysaber.get_trans_masks() is also useful to check if user expectations for the mask arrays are satisfied.
Example python scripts that demonstrate the above procedure are shown below. To obtain the data that is required to run this script, download and unzip the zip file at the link data. To run the script as is within the current working directory, the files in the zip file must be placed within a folder called data.

Verify transmission function and masks for horizontal edge radiograph.¶

import numpy as np
from PIL import Image #To read images in TIFF format
from pysaber import get_trans_masks #To compute transmission function and masks
import matplotlib.pyplot as plt #To display images
    
pix_wid = 0.675 #Width of each pixel in micrometers

#Read a horizontal edge radiograph for which the transmission function must be computed
rad = Image.open('data/horz_edge_25mm.tif') #Read radiograph
rad = np.asarray(rad) #Convert to numpy array
bright = Image.open('data/horz_bright.tif') #Read bright field
bright = np.asarray(bright) #Convert to numpy array
dark = Image.open('data/horz_dark.tif') #Read dark field
dark = np.asarray(dark) #Convert to numpy array
nrad = (rad-dark)/(bright-dark) #Normalize radiograph

#Get the transmission function and masks 
trans,trans_mask,rad_mask = get_trans_masks(nrad,edge='straight',pad=[3,3])
#Use pad = [1,1] if you do not want padding

#Display the array trans
#Visually check for inaccurate localization of the sharp-edge
plt.imshow(trans,cmap='gray')
plt.colorbar()
plt.show()

#Display and inspect the mask for transmission function
plt.imshow(trans_mask,cmap='gray')
plt.show()

#Display and inspect the mask for radiograph
plt.imshow(rad_mask,cmap='gray')
plt.show()

#Show a line plot comparing the measured radiograph and transmission function
sz = nrad.shape
coords = np.arange(-(sz[0]//2),sz[0]//2,1)*pix_wid
mid = (trans.shape[0]//2,trans.shape[1]//2)

plt.plot(coords,nrad[:,sz[1]//2])
#Due to padding, trans is three times the size of nrad in each dimension
#For proper alignment in the presence of padding, both nrad and trans are center aligned
#Center alignment is used since an equal amount of padding is applied at both ends of each axis
plt.plot(coords,trans[mid[0]-(sz[0]//2):mid[0]+(sz[0]//2),mid[1]])
plt.xlabel('micrometers')
plt.legend(['Measured','Transmission'])
plt.show()

Verify transmission function and masks for vertical edge radiograph.¶

import numpy as np
from PIL import Image #To read images in TIFF format
from pysaber import get_trans_masks #To compute transmission function and masks
import matplotlib.pyplot as plt #To display images
    
pix_wid = 0.675 #Width of each pixel in micrometers

#Read a vertical edge radiograph for which the transmission function must be computed
rad = Image.open('data/vert_edge_25mm.tif') #Read radiograph
rad = np.asarray(rad) #Convert to numpy array 
bright = Image.open('data/vert_bright.tif') #Read bright field
bright = np.asarray(bright) #Convert to numpy array
dark = Image.open('data/vert_dark.tif') #Read dark field
dark = np.asarray(dark) #Convert to numpy array
nrad = (rad-dark)/(bright-dark) #Normalize radiograph

#Get the transmission function and masks 
trans,trans_mask,rad_mask = get_trans_masks(nrad,edge='straight',pad=[3,3])
#Use pad = [1,1] if you do not want padding

#Display the array trans
#Visually check for inaccurate localization of the sharp-edge
plt.imshow(trans,cmap='gray')
plt.colorbar()
plt.show()

#Display and inspect the mask for transmission function
plt.imshow(trans_mask,cmap='gray')
plt.show()

#Display and inspect the mask for radiograph
plt.imshow(rad_mask,cmap='gray')
plt.show()

#Show a line plot comparing the measured radiograph and transmission function
sz = nrad.shape
coords = np.arange(-(sz[1]//2),sz[1]//2,1)*pix_wid
mid = (trans.shape[0]//2,trans.shape[1]//2)

plt.plot(coords,nrad[sz[0]//2,:])
#Due to padding, trans is three times the size of nrad in each dimension
#For proper alignment in the presence of padding, both nrad and trans are center aligned
#Center alignment is used since an equal amount of padding is applied at both ends of each axis
plt.plot(coords,trans[mid[0],mid[1]-(sz[1]//2):mid[1]+(sz[1]//2)])
plt.xlabel('micrometers')
plt.legend(['Measured','Transmission'])
plt.show()