A Comparison Study on Copy-Cover Image Forgery Detection

Sheng-Wen Peng
PCA Domain Method
Nowadays, due to rapid advances and
availabilities of powerful image processing
software, modifying the content of digital
images becomes much easier with the help of
sophisticated software such as Adobe
a photograph of Lincoln's head was
superimposed onto a portrait of the southern
leader John Calhoun
There are many ways to categorize the image
tampering, and generally, we can tell that some
usually performed operations in image tampering
1. Deleting or hiding a region in the image.
2. Adding a new object into the image
3. Misrepresenting the image information;
to insert and splicing image part of the original
image is the one of the most typical method.
Watermarking is the popular way to counterfeit
the forgery image.
Digital watermarking is the process of
embedding information into a digital signal
which may be used to verify its authenticity or
the identity of its owners. There are several
characteristic of watermarked image:
1. Robustness
2. Perceptibility
3. Capacity
4. Embedding method
The digital watermark, unlike the printed
visible stamp watermark, is designed to be
invisible to viewers. The bits embedded into an
image are scattered all around to avoid
identification or modification. Therefore, a
digital watermark must be robust enough to
survive the detection, compression, and
operations that are applied on.
However, watermarking techniques have some
drawbacks. Fragile watermark is not suitable
for such applications involving compression of
images, which is a common practice before
sharing images on the Internet.
The copy-cover technique is the most popular
technique for making image forgery. Copycover means that one portion of a given image
is copied and then used to cover some object in
the given image.
Several researchers have explored the copycover image forgery detection. Popescu and
Farid used the PCA domain representation to
detect the forged part, even when the copied
area is corrupted by noise.
Principal components analysis (PCA) is known
as the best data representation in the leastsquare sense for classical recognition [16]. It is
commonly used to reduce the dimensionality
of images and retain most information. The
idea is to find the orthogonal basis vectors or
the eigenvectors of the covariance matrix of a
set of images, with each image being treated as
a single point in a high dimensional space.
Xq = PX
For forgery detection, we create an image
database composed of 500 images for use in
our experiment.
1. Given image loading
 2. Convert the image given into gray scale
Let N be the total number of pixels
Let b denote the number of pixels in a square
Using PCA
 3. Find the eigenvectors and eigenvalues.
 4. Sort the eigenvalues in decreasing order and
also the eigenvectors.
 5. Compute the projections of the centered
testing gram matrix on the ordered
eigenvectors (decreasing order of eigenvalues).
6. Lexicographically sort the projected version of
the centered testing gram matrix. Let it be denoted
by S.
7. For every ith rows i in S, select a number of
subsequent rows, sj such that |i-j|<=Rth and place
all the pairs of coordinates (xi,yi) and (xj,yj) on to a
list P_in.
8. Compute offset for each row of P_in.
9. Compute the frequency offset.
10. Those rows in P_in which have high frequency
offsets are the duplicated regions
Frank Y. Shih* and Yuan Yuan ‘’A Comparison
Study on Copy-Cover Image Forgery
Detection’’ 2010
Digital Watermarking from Wikipedia
Principal component analysis from Wikipedia
Prem Kumar ‘’Digital Image Forensics (Copypaste forgery detection)’’ 2010

similar documents