Benjamin_Loison/Robust_image_source_identification_on_modern_smartphones
1
0
Fork 0
Code Issues 68 Pull Requests Packages Projects Releases Wiki Activity

Revert to some extent previous commit

Browse Source
This commit is contained in:
Benjamin Loison 2024-03-29 01:07:48 +01:00
parent 4382b3d649
commit d59a251b1f
Signed by: Benjamin_Loison
SSH Key Fingerprint: SHA256:BtnEgYTlHdOg1u+RmYcDE0mnfz1rhv5dSbQ2gyxW8B8
1 changed files with 7 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
datasets/noise_free_test_images/estimate_prnu.py
View File

@ -19,7 +19,7 @@ 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_FACTORS = [0.1, 0.01]
PRNU_FACTOR = 0.1
IMAGE_SIZE_SHAPE = (469, 704)
np.random.seed(0)
@ -39,31 +39,11 @@ for imageName in os.listdir(datasetPath):
imageWithoutPrnuPil = Image.open(imagePath).convert('F')
imageWithoutPrnuNpArray = np.array(imageWithoutPrnuPil)
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
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)
imagesPrnuEstimateNpArray += [imagePrnuEstimateNpArray]
plt.tight_layout(pad = 0)
plt.show()