1
Digital Image Forgery
Detection
Presentation Division
Digital Image Forgery Detection
Alteration
of the semantic components of a digital
image.
Removing
Adding
Types
Contents from the image
Data to the image
of Forgery
Image
Retouching
Image
Splicing (Copy-Paste)
Image
Cloning (Copy-Move)
4
Image Retouching
One of the oldest types of image forgery
Image features are tampered with.
Used to enhance or reduce digital image features.
Considered less dangerous type of image forgery.
Image Splicing (Copy-Paste)
Fragments of 2 or more images are combined to form an image.
This operation is fundamental in digital photo montaging and in turn is a
mechanism for image forgery creation.
Image splicing technique may change the visual message of digital images
more aggressively than image retouching.
Image Cloning (Copy-Move)
Considered as a special case of image splicing, where the tampering occurs
within a single image and no need for multiple images.
Part of the image is copied and then pasted in a desired location within the
same image.
The purpose of such tampering is to duplicate or conceal a certain object in
that image.
Image Cloning
Blurring is usually used to reduce the expected irregularity along the border
of the pasted regions.
The similarity of texture, color, noise and other information inside the image
make it very difficult to detect this kind of tampering via visual inspection.
Moreover, performing of post-processing operations such as blurring, adding
noise and JPEG compression or geometric operations such as scaling, shifting
and rotation increase the hardness of detection task.
Forgery Detection Mechanisms
Can be Classified into Two Types
Active Methods
Passive Methods
Active Methods
Hidden Information inside the Digital Image.
Done at the time of Data Acquisition or before disseminated
to the public.
Embedded information can be used to identify the source of
such image or to detect possible modification to that image.
Forgery Detection Mechanisms
(Active Methods)
Two Major Types
Digital Signature
Digital Watermarking
10
Forgery Detection Mechanisms
(Passive Methods)
Use traces left by the processing steps in different phases of acquisition and
storage of digital images.
These traces can be treated as a fingerprint of the image source device.
Passive methods work in the absence of protecting techniques.
They do not use any pre-image distribution information inserted into digital
image.
They work by analyzing the binary information of digital image in order to
detect forgery traces, if any
Limitation is the number of false positives.
11
High Precision Rotation Angle Estimation for Rotated
Images
Paper addresses the detection of copy-move(cloning) technique
As discussed before cloning detection becomes harder when the forger uses
geometric alterations like scaling, rotation & shifting.
Particularly addresses the Rotation transformation.
This paper proposes a novel blind image rotation detection algorithm with
high precision rotation angle estimation
12
High Precision Rotation Angle Estimation for Rotated
Images
I= Original image
I=Intermediate Image
I= Rotated Image
S, t=pixel coordinates in the rotated image I.
= weighted value.
13
High Precision Rotation Angle Estimation for Rotated
Images
R and S are constant(translation)
= horizontal distance rotated image I & intermediate image I
= vertical distance.
14
High Precision Rotation Angle Estimation for Rotated
Images
For a single pixel, we have:
15
High Precision Rotation Angle Estimation for Rotated
Images
Plot of horizontal distance vector and
its spectrum at
Plot of peak frequency of distance vector
against all . Frequency is normalized to.
16
High Precision Rotation Angle Estimation for Rotated
Images
Algorithm of the approach
17
High Precision Rotation Angle Estimation for
Rotated Images
Resolution
Total Images
Correct Images
Correct Rate
500
486
97.2%
500
480
96.0%
500
471
94.2%
500
459
91.8%
500
438
87.6%
18
High Precision Rotation Angle Estimation for Rotated
Images
Experiment
results. 1st column: three images rotated at respectively; 2nd column: theoretical
19
pixel variance spectrum for the rotated images; 3rd column: actual pixel variance spectrum
for
the rotated images.
High Precision Rotation Angle Estimation for Rotated
Images
Conclusion
In this paper, propose a blind image rotation angle estimation
method is proposed by exploring the periodicity of pixel variance
of rotated images.
Experiment results show that this method works well for rotation
angles larger than , but not as good for smaller rotation angles.
The method can be used in areas like copy-paste image forgery
detection. In the future, the author plans to modify the algorithm
to improve the correct rate of small rotation angle estimation.
20
Region Duplication Forgery Detection
using Hybrid Wavelet Transforms
21
Region Duplication Forgery Detection using Hybrid
Wavelet Transforms
Discrete Cosine Transforms
Starts by dividing the MN suspicious image into small overlapping blocks.
This step is achieved by sliding a window of size BB from the upper left
corner to the lower right corner one pixel each time.
The quantized DCT coefficients are extracted from each block and used to
represent the features of these blocks.
The quantized DCT coefficients are stored as one row in a matrix A of (M-B+1)
(N-B+1) rows and B B columns, where B B is the block size.
Two identical rows in the matrix A, correspond to two identical blocks in the
suspicious image.
22
Region Duplication Forgery Detection
using Hybrid Wavelet Transforms
Hadamard Walsh Transforms
The Product of a Boolean Function and a Walsh Matrix is a Walsh Spectrum
23
Region Duplication Forgery Detection
using Hybrid Wavelet Transforms
Example of Copy-Move Forgery, (a) Original Image (b) Forged Image
24
Thank you
25
