Use np.clip

6f961a46cf

datasets/noise_free_test_images/estimate_prnu.py

@@ -24,7 +24,7 @@ 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 = 0.01
-NOISE_FACTOR = 0.1
+NOISE_FACTOR = 0.2
 
 np.random.seed(0)
 
@@ -77,13 +77,13 @@ for splitNXNIndex, splitNXN in enumerate(SPLIT_N_X_N_S):
                     axis[1].set_title('First image Gaussian noise')
                     axisImShow(axis[1], imageNoise)
 
-                    axis[2].set_title('First image with Gaussian noise')
+                    axis[2].set_title('First image with Gaussian noise\nClipped between 0 and 255')
                     axisImShow(axis[2], np.clip(imageWithoutPrnuNpArray + imageNoise, 0, 255))
 
-                    axis[3].set_title('Actual Gaussian noised PRNU')
-                    axisImShow(axis[3], prnuNpArray)
+                    axis[3].set_title('Actual Gaussian noised PRNU\nClipped between -1 and 1')
+                    axisImShow(axis[3], np.clip(prnuNpArray, -1, 1))
 
-                    axes[1][0].set_title('First image with Gaussian noise and PRNU')
+                    axes[1][0].set_title('First image with Gaussian noise and PRNU\nClipped between 0 and 255')
                     axisImShow(axes[1][0], np.clip(imageWithoutPrnuNpArray + prnuNpArray + imageNoise, 0, 255))
                 isFirstImage = False