Add PRNU_FACTOR 0.1 and 0.01 view on a single figure

algorithms/context_adaptive_interpolator/context_adaptive_interpolator.py

@@ -6,6 +6,7 @@ from wiener_filter import wienerFilter
 
 # Assume greyscale PIL image passed.
 # What about other color channels? See #11.
+# `THRESHOLD` seems to have been designed to assume 256 range based images.
 def contextAdaptiveInterpolator(I, IImage, showProgress = False):
     rImage = Image.new('L', (IImage.size[0] - 2, IImage.size[1] - 2))
     r = rImage.load()

datasets/noise_free_test_images/estimate_prnu.py

@@ -3,6 +3,7 @@
 import os
 from PIL import Image
 import numpy as np
+import matplotlib.pyplot as plt
 import sys
 
 sys.path.insert(0, '../../algorithms/image_utils/')
@@ -13,34 +14,56 @@ sys.path.insert(0, '../../algorithms/context_adaptive_interpolator/')
 
 from context_adaptive_interpolator import contextAdaptiveInterpolator
 
+from skimage.restoration import denoise_tv_chambolle
+
 datasetPath = 'no_noise_images'
 # Note that contrarily to `datasets/fake/`, here we do not have images being Gaussian with `scale` `1` but actual images with pixel values between 0 and 255.
 # In addition to the range difference, note that the distribution in the first set of images was a Gaussian and here is very different and specific.
-PRNU_FACTOR = 0.15
+PRNU_FACTORS = [0.1, 0.01]
 IMAGE_SIZE_SHAPE = (469, 704)
 
 np.random.seed(0)
 
 #prnuNpArray = 255 * randomGaussianImage(scale = PRNU_FACTOR, size = IMAGE_SIZE_SHAPE)
 prnuPil = Image.open('prnu.png').convert('F')
-prnuNpArray = np.array(prnuPil) * PRNU_FACTOR
+prnusNpArray = [np.array(prnuPil) * PRNU_FACTOR for PRNU_FACTOR in PRNU_FACTORS]
 
 def isIn256Range(x):
     return 0 <= x and x <= 255
 
+imagesPrnuEstimateNpArray = []
+
 for imageName in os.listdir(datasetPath):
     if imageName.endswith('.png'):
         imagePath = f'{datasetPath}/{imageName}'
         imageWithoutPrnuPil = Image.open(imagePath).convert('F')
         imageWithoutPrnuNpArray = np.array(imageWithoutPrnuPil)
-        #showImageWithMatplotlib(imageWithoutPrnuNpArray)
-        imageWithPrnuNpArray = imageWithoutPrnuNpArray + prnuNpArray
-        showImageWithMatplotlib(imageWithPrnuNpArray)
-        break
-        assert all([isIn256Range(extreme) for extreme in [imageWithPrnuNpArray.max(), imageWithPrnuNpArray.min()]]), 'Adding the PRNU resulted in out of 256 bounds image'
-        imageWithPrnuPil = toPilImage(imageWithPrnuNpArray)
-        imageWithPrnuCaiPil = contextAdaptiveInterpolator(imageWithPrnuPil.load(), imageWithPrnuPil)
-        imageWithPrnuCaiNpArray = np.array(imageWithPrnuCaiPil)
-        showImageWithMatplotlib(imageWithPrnuCaiNpArray)
+
+        fig, axes = plt.subplots(3, 2)
+        fig.suptitle('Single PRNU estimation from an image with PRNU')
+
+        axes[0][0].set_title('Actual PRNU')
+        axes[0][0].imshow(prnuNpArray)
+
+        axes[0][1].axis('off')
+
+        for prnuIndex, prnuNpArray in enumerate(prnusNpArray):
+            imageWithPrnuNpArray = imageWithoutPrnuNpArray + prnuNpArray
+            #assert all([isIn256Range(extreme) for extreme in [imageWithPrnuNpArray.max(), imageWithPrnuNpArray.min()]]), 'Adding the PRNU resulted in out of 256 bounds image'
+            imageWithPrnuPil = toPilImage(imageWithPrnuNpArray)
+            #imagePrnuEstimatePil = contextAdaptiveInterpolator(imageWithPrnuPil.load(), imageWithPrnuPil)
+            #imagePrnuEstimateNpArray = np.array(imagePrnuEstimatePil)
+            imagePrnuEstimateNpArray = imageWithPrnuNpArray - denoise_tv_chambolle(imageWithPrnuNpArray, weight=0.2, channel_axis=-1)
+            axis = axes[prnuIndex + 1]
+
+            axis[0].set_title(f'Image with PRNU\nPRNU_FACTOR = {PRNU_FACTORS[prnuIndex]}')
+            axis[0].imshow(imageWithPrnuNpArray)
+
+            axis[1].set_title('PRNU estimate')
+            axis[1].imshow(imagePrnuEstimateNpArray)
         break
 
+        imagesPrnuEstimateNpArray += [imagePrnuEstimateNpArray]
+
+plt.tight_layout(pad = 0)
+plt.show()