Overview Repositories
65
 Projects Packages Public Activity Starred Repositories
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#25 2024-03-29 01:48:12 +01:00
Estimating fake PRNU on noise-free images

Let us not consider this RMS inconsistency but admit that the visual clear PRNU is fine, so let us add Gaussian noise.

2f66e82f22 Show both RMS and RMS normalized
3ccec5bbd0 Add RMS computation
f297060f42 Split images in 4x4 to increase PRNU estimation accuracy
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#25 2024-03-29 01:15:17 +01:00
Estimating fake PRNU on noise-free images

prnu_4x4

9b57d3441c Render PRNU estimate taking into account all images
d59a251b1f Revert to some extent previous commit
4382b3d649 Add PRNU_FACTOR 0.1 and 0.01 view on a single figure
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#25 2024-03-29 00:43:02 +01:00
Estimating fake PRNU on noise-free images

PRNU quite guessable on a given image. Now need to mean to make it clear.

Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#25 2024-03-29 00:34:14 +01:00
Estimating fake PRNU on noise-free images

As Context-Adaptive Interpolator is slow and possibly incorrectly implemented, let us try a good known working denoiser:

Figure_1

…
dfe2540c02 Apply Context-Adaptive Interpolator
9a3cfd7ba1 Add PRNU showing such
ba5a1b742b #24: Make executions reproducible
Generate arbitrary shape PRNU shown as such
What do phone PRNU distribution looks like?
a99e942d3a Add datasets/noise_free_test_images/estimate_prnu.py
0953fb7475 Add issue reference to datasets/fake/generate_dataset.py
Compare 2 commits »
64d876c0fc Add datasets/noise_free_test_images/estimate_prnu.py
82e7026264 Add algorithms/image_utils/image_utils.py and move there randomGaussianImage and showImageWithMatplotlib
70ccb094d5 Rename randomImage to randomGaussianImage
Benjamin_Loison commented on issue Benjamin_Loison/Robust_image_source_identifica…#19 2024-03-28 16:07:06 +01:00
Theoretical correct Gaussian noise image?

https://pillow.readthedocs.io/en/latest/handbook/concepts.html#modes