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

#63: Add debugging

Browse Source
This commit is contained in:
Benjamin Loison 2024-04-30 06:45:28 +02:00
parent a8725e5e88
commit 7a807b91d8
No known key found for this signature in database
1 changed files with 29 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
datasets/raise/attribute_source_camera.py
View File

@ -27,15 +27,14 @@ random.seed(0)
for camera in IMAGES_CAMERAS_FOLDER:
random.shuffle(imagesCamerasFileNames[camera])
minimumNumberOfImagesCameras = 4#min([len(imagesCamerasFileNames[camera]) for camera in IMAGES_CAMERAS_FOLDER])
minimumNumberOfImagesCameras = 16#min([len(imagesCamerasFileNames[camera]) for camera in IMAGES_CAMERAS_FOLDER])
for camera in IMAGES_CAMERAS_FOLDER:
imagesCamerasFileNames[camera] = imagesCamerasFileNames[camera][:minimumNumberOfImagesCameras]
imagesCamerasFileNames[camera] = imagesCamerasFileNames[camera][:minimumNumberOfImagesCameras]#[imagesCamerasFileNames[camera][0]] * minimumNumberOfImagesCameras#[:minimumNumberOfImagesCameras]
print(camera, imagesCamerasFileNames[camera])
numberOfCameras = len(IMAGES_CAMERAS_FOLDER)
camerasIterativeMean = {camera: iterativeMean() for camera in IMAGES_CAMERAS_FOLDER}
minColor = None
maxColor = None
# Assume that for each camera, its images have the same resolution.
# The following consider a given color channel resolution, assuming they all have the same resolution.
minimalColorChannelCameraResolution = None
@ -46,14 +45,17 @@ for camera in IMAGES_CAMERAS_FOLDER:
if minimalColorChannelCameraResolution is None or singleColorChannelImagesShape < minimalColorChannelCameraResolution:
minimalColorChannelCameraResolution = singleColorChannelImagesShape
minColor = 13#None
maxColor = 7497#None
minColor = None#13#None
maxColor = None#7497#None
accuracy = []
numberOfTrainingImages = int(minimumNumberOfImagesCameras * TRAINING_PORTION)
numberOfTestingImages = minimumNumberOfImagesCameras - int(minimumNumberOfImagesCameras * TRAINING_PORTION)
cameraTestingImagesNoise = {}
from utils import silentTqdm
#tqdm = silentTqdm
returnSingleColorChannelImage = lambda singleColorChannelImage, _minColor, _maxColor: singleColorChannelImage
for computeExtremes in tqdm(([True] if minColor is None or maxColor is None else []) + [False], 'Compute extremes'):
@ -66,6 +68,7 @@ for computeExtremes in tqdm(([True] if minColor is None or maxColor is None else
# Should make a function
imageFileName = imagesCamerasFileNames[camera][numberOfTrainingImages + cameraTestingImageIndex]
imageFilePath = f'{IMAGES_CAMERAS_FOLDER[camera]}/{imageFileName}'
print(f'{imageFilePath=}')
# Should make a function
singleColorChannelImages = {color: rescaleIfNeeded(getColorChannel(imageFilePath, color)[:minimalColorChannelCameraResolution[0],:minimalColorChannelCameraResolution[1]], minColor, maxColor) for color in Color}
@ -74,7 +77,7 @@ for computeExtremes in tqdm(([True] if minColor is None or maxColor is None else
multipleColorsDenoisedImage = mergeSingleColorChannelImagesAccordingToBayerFilter(singleColorChannelDenoisedImages)
imagePrnuEstimateNpArray = multipleColorsImage - multipleColorsDenoisedImage
cameraTestingImagesNoise[camera] = cameraTestingImagesNoise.get(camera, []) + [multipleColorsDenoisedImage]
cameraTestingImagesNoise[camera] = cameraTestingImagesNoise.get(camera, []) + [imagePrnuEstimateNpArray]#multipleColorsDenoisedImage]
for cameraTrainingImageIndex in tqdm(range(minimumNumberOfImagesCameras if computeExtremes else numberOfTrainingImages), 'Camera training image index'):
for cameraIndex, camera in enumerate(tqdm(IMAGES_CAMERAS_FOLDER, 'Camera')):
imageFileName = imagesCamerasFileNames[camera][cameraTrainingImageIndex]
@ -90,10 +93,16 @@ for computeExtremes in tqdm(([True] if minColor is None or maxColor is None else
multipleColorsDenoisedImage = mergeSingleColorChannelImagesAccordingToBayerFilter(singleColorChannelDenoisedImages)
imagePrnuEstimateNpArray = multipleColorsImage - multipleColorsDenoisedImage
#print(camera)
#plt.imshow(multipleColorsImage)
#plt.show()
#exit(1)
cameraIterativeMean = camerasIterativeMean[camera]
cameraIterativeMean.add(imagePrnuEstimateNpArray)
plt.imsave(f'm_{escapeFilePath(camera)}.png', camerasIterativeMean[camera].mean)
if cameraIndex == numberOfCameras - 1:
numberOfTrainingImagesAccuracy = 0
print(f'{numberOfTestingImages=} {numberOfCameras=}')
# Loop over each camera testing image folder.
for actualCamera in IMAGES_CAMERAS_FOLDER:
for cameraTestingImageIndex in tqdm(range(numberOfTestingImages), 'Camera testing image index'):
@ -101,11 +110,23 @@ for computeExtremes in tqdm(([True] if minColor is None or maxColor is None else
minimalDistance = None
# Loop over each camera to compute closeness between the considered testing image noise and the estimated PRNUs of the various cameras.
for camera in IMAGES_CAMERAS_FOLDER:
distance = rmsDiffNumpy(cameraTestingImagesNoise[camera][cameraTestingImageIndex], camerasIterativeMean[camera].mean)
distance = rmsDiffNumpy(cameraTestingImagesNoise[actualCamera][cameraTestingImageIndex], camerasIterativeMean[camera].mean)
'''
print(f'{camerasIterativeMean[camera].numberOfElementsInMean=}')
print(f'{cameraTestingImagesNoise[actualCamera][cameraTestingImageIndex].min()=}')
print(f'{cameraTestingImagesNoise[actualCamera][cameraTestingImageIndex].max()=}')
print(f'{camerasIterativeMean[camera].mean.min()=}')
print(f'{camerasIterativeMean[camera].mean.max()=}')
plt.imsave(f'a_{actualCamera}.png', cameraTestingImagesNoise[actualCamera][cameraTestingImageIndex])
plt.imsave(f'b_{camera}.png', camerasIterativeMean[camera].mean)
plt.show()
'''
print(f'{cameraTestingImageIndex=} {camera=} {actualCamera=} {distance=}')
#exit(1)
if minimalDistance is None or distance < minimalDistance:
minimalDistance = distance
cameraPredicted = camera
print(f'Predicted camera {cameraPredicted} {"good" if cameraPredicted == actualCamera else "bad"}')
if cameraPredicted == actualCamera:
numberOfTrainingImagesAccuracy += 1
accuracy += [numberOfTrainingImagesAccuracy / (numberOfTestingImages * numberOfCameras)]