Exposing Digital Forgeries in Color Array Interpolated Images

1
Exposing Digital Forgeries in Color Array
Interpolated Images

• Presented by
• Ariel Hutterer

Final Fantasy ,2001
My eye
2
References

• Alin C.Popescu and Hany Farid
• Exposing Digital Forgeries in Color Filter Array
  Interpolated Images.
• Yizhen Huang
• Can Digital Forgery Detection Unevadable? A
  Case Study Color Filter Array Interpolation
  Statistical Feature Recovery.
• Hagit El Or
• Demosaicing.

3
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Crack Methods
• Computer Graphics

4
Introduction- forgeries

• Low cost cameras ,photo editing software.
• Images can be manipulated easily.
• Splicing.

5
Introduction- forgeries

• Images have a huge impact in public opinion.
• Legal world.
• Scientific evidence.

6
Introduction - preventing forgeries approaches

• Two principal approaches to prevent forgeries
• Digital watermarking
• Means that image can be authenticated.
• Drawbacks
• Specially equipped digital cameras ,that insert
  the watermark.
• Assume that watermark cannot be easily removed
  and reinserted. (but .it is???)
• Statistic analysis
• Most color digital cameras , introduces specific
  correlation
• A third of the image are captured by a sensor.
• Two thirds of the image are interpolated.
• Images manipulated must alter this specific
  statistic.

7
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Crack Methods
• Computer Graphics

8
Digital Cameras

• Most Color digital Cameras have a single
  monochrome Array of sensors

9
Digital Cameras

• How does color form with monochrome sensor for
  each pixel?

10
Digital Cameras-Bayer Color Array

• Half pixels are Green ,quarter are Red and
  quarter are Blue

11
Digital Cameras-Bayer Color Array

• Several possible arranges

Diagonal Bayer
Bayer
Diagonal
Striped
Psudo-random Bayer
12
Digital cameras - forming color
13
Digital cameras - forming color
14
Digital cameras - forming color
Interpolation
15
Digital cameras - forming color

• Bayer Array For almost all Digital Cameras
• Color Interpolation different for each make of
  Digital Camera

Interpolation
16
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Crack Methods
• Computer Graphics

17
Interpolations

• Naive per channel interpolation
• Nearest neighbor ,Bilinear interpolation
• Inter-channel dependencies and correlations
• Reconstruct G channel, then reconstruct R B
  based on G. Reconstruct all 3 channels
  constrained with inter-channel dependence.
• Adaptive reconstruction
• Measure local image variations (e.g. edges,
  gradients, business) and reconstruct accordingly.

18
Interpolations - Aliasing
Interpolate
19
Interpolations - Aliasing
Result
Interpolate
20
Interpolations - Samples
21
Interpolation-Bilinear Bicubic

• Red and Blue Kernels
• Separable 1-D filters

Rw
Rw ½(RnwRsw)
22
Interpolation-Bilinear Bicubic

• Green kernels
• 2-D filters

23
Interpolation- Gradient Based

• First, calculate Green channel
• Calculate derivates estimators
• Determination of Greens values

24
Interpolation Evaluation Tools
25
Interpolation -Results
Original
Linear
Kimmel
26
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Cracks Methods
• Computers Graphics

27
Detecting CFA Interpolation

• In Each pixel only one color derives from the
  sensor ,two others derive from interpolation from
  their neighbors .
• The correlation are periodic.
• Tampering will destroy these correlations.
• Splicing together two images from different
  cameras will create inconsistent correlations
  across the composite image.

28
Detecting CFA Interpolation

• Two different tools
• EM algorithm
• Produce Map of Probabilities and interpolation
  coefficients
• Used to detect kind of interpolation
• Farids Indicator
• Produce Map of Similarities
• Used to quantify the similarity to CFA
  Interpolated Image

29
EM Algorithm (Expectation/Maximization)

• Two possible models
• M1the sample is linearly correlated to its
  neighbors
• M2the sample is not correlated to its neighbors

30
EM Algorithm (Expectation/Maximization)

• f(x,y) color channel
• alpha - parameters ,where(0,0) 0. denotes
  the specific correlation.
• n - independent and identically samples
  drawn from a Gaussian distribution, with 0 mean
  and unknown variance

31
EM Algorithm (Expectation/Maximization)

• Two-step iterative algorithm
• E-step calculate the probability of each sample
• M-step the specific form of the correlation is
  estimated.
• Based in Bayes rule

32
Farids indicator

• The similarity between the probability and a
  synthetic map is obtained by
• Where
• Similarity measure is phase insensitive

33
Farids indicator

• How to use it
• CFA-Interpolated if at least one channel is
  greater than threshold1
• Non CFA Interpolated if all 3 channels are
  smaller than threshold2

result
threshold2
threshold1
CFA Interpolated
Non CFA Interpolated
Unknown
Ind(cfa-sf)
Ind(cfa-isf)
34
Huang indicator

• Motivation Farids Indicator is proportional to
  image size.
• Table of Green Channel Indicator
• Huang Indicator

Indicator function 32x32 128x128 256x256 512x512
Farid 140 2303 9419 52361
Huang 2.70 2.70 2.84 4.31
35
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Cracks Methods
• Computers Graphics

36
Results

• Detecting different interpolation methods
• Detecting tampering
• Measuring Sensitivity and robustness

37
Detecting different interpolation methods

• Hundreds of images from 2 digital cameras
• Blur 3x3
• Down sampled
• Cropped
• Resample in CFA Interpolations

38
Detecting different interpolation methods
39
Detecting different interpolation methods
40
Detecting different interpolation methods
41
Detecting different interpolation methods
42
Detecting different interpolation methods

• Coefficients are 8 to each color so we are a 24-D
  vector ,LDA classifier ,results
• 97 Interpolations kinds was detected
• 2D projection of LDA

43
Detecting tampering

• Hiding the damage of the car
• Air-brushing ,smudging ,blurring and duplication

44
Detecting tampering

• Result
• Left F(p) for tampered portion
• Right F(p) for unadulterated portion

45
Measuring Sensitivity and robustness

• Testing different interpolations with Farids
  indicator

False 0 Median 5x5 97
Bilinear 100 Gradient based 100
Bicubic 100 Adaptive color plane 97
Median 3x3 99 Variable number of gradients 100
remember
46
Measuring Sensitivity and robustness

• Testing influence of jpeg

47
Measuring Sensitivity and robustness

• Testing influence of Gaussian Noise

48
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Crack Methods
• Computer Graphics

49
Cracking

• Whats a true digital image
• General Model

50
True digital image

• It was taken by a CCD/CMOS digital camera, or
  other device with similar function and remains
  intact after shooting except for embedding
  ownership and other routinely added information.

51
General Model

• where
• W all images
• S all possible images tacked by an ideal camera
  c.
• N are S enlarged because of noise.
• Detection method
• Pm(I), a projection of Image I
• I is true when
• I is Artificially CFA-interpolated

52
General Model

• The result image should be as close as possible
  to the original
• The mean of the difference to an ideally CFA
  interpolated image should be controlled in a
  specific range.
• Such difference should be distributed averagely.

53
General Model

• Im Tampered Image
• Im cracked Image
• Int(I) Ideal Interpolated

Dif(Im,Im,Int(Im))
K2
K1
Dif(Im,Im)
Dif(Im,Int(Im))
54
General Model

• We are looking for
• We want to minimize the 3d distance

55
Outline

• Introduction
• Digital Cameras
• Interpolations
• Detecting CFA Interpolation
• Results
• Crack Methods
• Computer Graphics

56
Computer Graphic

• A naïve approach
• Computer Graphic will be detected like non
  CFA-Interpolated.

57
Computer Graphics

• Huge improvement of dedicated hardware in the
  last 7 years
• SGIOnyx2 ,Infinity reality 3(2000)
• 12 bits 4 channels
• No shaders
• End User license ,250,000
• Pc d/core, geforce 8(2006)
• 32 bits 4 channels
• Shaders w/24 parallels pipes
• 1,500-5,000

58
Computer Graphics

• 2001,Final fantasy ,first Film made with PC.

59
Computer Graphics

• See cg not like an Image, see it like REALITY.

Render Reality high resolution ,by 32bits for
each color
Optical distortions, ghost and blurring
Sensor CFA sampling and noise
Interpolation
Image
60
Computer Graphics From Image Forgeries to
Science Fiction

• Image forgeries are a positive issue for
  development of
• Simulators.
• Trainers.
• Robots

61
Computer Graphics From Image Forgeries to
Science Fiction
62
Conclusion

• Detection CFA-Interpolated methods are not enough
  robust.
• Compression like jpeg destroy the interpolation
  correlation.
• Interpolation can be artificially made.

The End