Copy-Move Forgery Detection Using DCT

International Journal of Emerging Technologies and Engineering (IJETE)
Volume 2 Issue 3, March 2015, ISSN 2348 8050
COPY-MOVE FORGERY DETECTION USING DCT

Neha Jadhav1, Suvarna Kharat2, Punam Nangare3
Nutan Maharashtra Institute Of Engineering And Technology, Pune
ABSTRACT:
In today's day today life digital images
are available everywhere and it is very easy to
manipulate these digital images by using
powerful editing software. Now a day's many
people add, crop or remove important features
from an image without leaving any proof of fake
images. There are many techniques used for
forgery detection. One of the technique most
commonly used is Copy-Move forgery in which
coping a some part of image and pasting it into
the same image in order to hide some data or
part of an image and other most commonly used
technique is staganalysis in which some message
is hidden inside the image which is not easily
possible to see with naked human eye. In this
paper we search the problem of detecting the
forgery and describe robust detection method.
this method successfully detect the forged part
even when the copied area is edited to combine
it with the background of an image and even if
the forged image is saved in the JPEG format.
KEYWORDS: Digital image , stenography
,Copy-Move
I.
Forgery detection techniques divided

into two major categories: active and passive
methods.Active method requires some prior
information of an image hence such methods are
not useful while handling images from unknown
sources.this is biggest drawback of active
method.digital watermarking belongs to active
method.Passive method does not require any
prior information of digital image. The method
works purely by analyzing binary information of
digital
image
without
any
external
information.Copy-move forgery belongs to this
method. Some examples of forgery is shown
below.
(a)
(b)
Fig1.An example of copy-move forgery:(a) the
forged image with four missiles & (b) the
original image with three missiles.
INTRODUCTION:
Now a days, digital images are widely

used all over the world .From newspapers to
magazines, fashion industry, in medical field,
science field, forensic labs etc. depends on
digital images. Exchanging soft copy of various
documents is a normal practice in these days. So
there is a possibility of forgery while exchanging
such type of documents .Forgery is manipulation
of an image to achieve a specific result. Image
Forgery is the process of making illegal changes
of image information. Forgery may occur in
every application which is using digital image
because user can change it by using editing tools
available in market.
Fig.2.Another example of typical copy-move

forgery .Left : the original image. Right : the
tempered image.
II.
PREVIOUS SYSTEM:
38
www.ijete.org

There are many techniques have been

proposed to solve the problem of forgeryas
given below:
[1]Bin YANG 1, Xingming SUN 2 proposed
robust and more reliable forgery method for
copy- move forgery detection based on DWTFWHT.They uses DWT to reduce the size of
image and expands the feature in the
overlapping blocks by FWHT instead of the
DCT. Their proposed method is weak in
detecting images which have undergone the
attack of transforming.
[2] Nathalie Diane Wandji1, Sun Xingming
proposed an DCT algorithm that can detect copy
move forgeries. They use the category of passive
methods because it does not require any primary
information of an image to perform operations.
Also it can detect multiple copy- move forgeries
in the same image and also it is relatively very
powerful to some common distortions.
[3] Ashima Gupta1, Nisheeth Saxena2, S.K
Vasistha3 proved that the use of DCT is better
than using PCA for detecting copy-move attacks
in highly textured images.They improve the
detection rate and time of the copy-move attack
proposed by their algorithm.
[4] Andrew D. Ker publish peper on
"Steganalysis of LSB matching in gray scale
images.They have given two ways to apply the
histogram characteristic function
in the
diagnosis of stegnography. They produce more
reliable
detectors
for
LSB
matching
steganography in grayscale images.
[5] Fridrich et al.firstly, divided an image into
overlapping blocks of equal size. The coefficient
of each block was removed by discrete cosine
transform. Finally, the duplicated regions were
found by matching the quantized coefficients
which had been simultaneously sorted.
[6] Popescu and Farid proposed a copy-move
forgery detection method in which the authors
extracted the coefficient by principal component
analysis (PCA) instead of DCT.Therefore the
dimensions of the coefficients extracted by PCA
are relatively smaller than DCT. The weakness
of this peper is that it cannot detect the rotating
copy region of an image.
III.
PROPOSED SYSTEM:
Image Forgery is the process of making

an illegal modification or reproduction of an
image information.In this paper,we used DCT
algorithm for forgery detection. There is an
approach that can detect tempered JPEG images
and further locate the tempered parts, by
observing the double quantization effect hidden
among the DCT coefficients.The JPEG process
is a widely used form of lossy image
compression that only used by the DCT.The
JPEG method is used for both Black & white
and colour images.
Our method detects
duplicated forgery regions by dividing the image
into 8*8 overlapping blocks and then we search
for the matching region in the image. We show
the potential of this method
on capable
forgeries and evaluate its strongness also.
A.BLOCK DIAGRAM:
Fig 3: Block diagram of DCT algorithm

B) WORKING OF DCT ALGORITHM:
1. Take tampered image as a input.
2. Divide input image into 8*8 block of pixels.
3. Apply DCT to each block of pixels.
4. After applying DCT each block is compressed
through quantization.
5. The array of compressed blocks that form
the image is stored in a drastic manner reduced
amount of space.
6. When desired, the image is rebuild through
decompression , a process that uses IDCT.
39
www.ijete.org

C)
CALCULATIONS
OF
DCT
ALGORITHM:
The DCT Equations:
The DCT equation represents the I,Jth entry of
the DCT of an image.
..(1)
To get the matrix form of equation (1), we will
use the following equations.
1
= 0
, =
2
2 +1
cos
> 0
ranging from 1 to 100 where 1 gives lower

image quality and higher compression while 100
gives better quality but lower compression.The
quality level Q50 matrix gives both high
compression and best decompression image
quality.
For quantization, it is obtained by

dividing each element in D matrix by
corresponding pixel values in the Q50 matix and
round up to the nearest integer value.
For an 8*8 block it results in following standard

matrix:-
, = round
.
,
E)CODING:
The quantized matrix C is now used for
final step of compression.In this step all
coefficients of C are converted into binary
stream by using encoder.After quantization most
of the coefficients results into zero.JPEG encode
this Quantized coefficients in the zig-zag
mannar as shown in figure3.
Firstly,we start with 8*8 block of image pixel
values that is selected from very uppermost left
side corner of an image.For DCT, pixel values
ranging from -128 to 127.Therefore from each
pixel values of 8*8 blocks 128 gets
substracted.After substraction,this result is
stored into some alphabet lets say M. Now, we
perform the DCT which is completed
successfully by matrix multiplication.
D=T*M*T
D) QUANTIZATION
Now, above DCT matrix is ready for
compression by quantization. In this step, by
selecting a specific quantization matrices it is
possible to vary the levels of image compression
and quality of an image.The image quality levels
Fig.4.process of coding
F) DE-COMPRESSION:
40
www.ijete.org

Decompression is exactly opposite

process of compression.Original
image is
obtained by decoding the bit stream of quantised
matrix C and multiplied by quantization matrix
Q50
R(I,j)= Q(I,j) * c(I,j)
For IDCT, 128 is added to each element
of matrix R which is rounded.Formula is given
below.
N=round(T * R * T) + 128
FLOWCHART OF DCT:
IV.
V.
RESULTS:
CONCLUSION:
Copy-move forgery is one of the most

commonly used forgery technique. In this we
use a stongest method to detect the tempered
region in an image. We have taken some test on
the algorithm against forge images from the
41
www.ijete.org

internet. From this algorithm, we get

improvements in the detection rate and the
detection time of copy move attack on images.
For highly textured images DCT is better than
any other forgery detection algorithms.
By using wavelet transform, we can
improve the efficiency of algorithm. For future
work, we will improve the accuracy of algorithm
and try to perform this process on video clips.
REFERANCES:
1) Tao Jing Xinghua li, Feifei Zhang,
Image Tamper Detection Algorithm
Based on Radon and fourier-Mellin
Transform,pp 212-215 IEEE2010
2) Sarah A. Summers, Sarah C.
WahlMultimedia Security and Forensic
Authentication of Digital images,
http://cs.uccs.edu/~cs525/studentproj/p
roj52006/sasummer/doc/cs525projsumm
ersWahl.doc.
3) J. Fridrich, D. Soukal, and J. Lukas,
Detection of Copy-Move Forgery in
Digital Images, in Proceedings of
Digital Forensic Research Workshop,
August 2003.
4) A. C. Popescu and H. Farid, Exposing
Digital
Forgeries
by
Detecting
Duplicated Image Regions, Technical
Report, TR2004-515, Department of
Computer Science, Dartmouth College,
pp. 758-767, 2006.
5) X. Kang and S. Wei, Identifying
Tampered Regions Using Singular
Value Decomposition in Digital Image
Forensics, International Conference on
Computer Science and Software
Engineering, pp. 926-930,2008
6) B. Mahdian and S. Saic, Detection of
copy-move forgery using a method
based on blur moment invariants.,
Elsevier Forensic Science International,
vol. 171, no. 2-3, pp. 180-189 Sep.
2007..
7) S.-jin Ryu, M.-jeong Lee, and H.-kyu
Lee, Detection of Copy-Rotate- Move
Forgery Using Zernike Moments, IH ,
LNCS 6387, vol. 1, pp. 51-65, 2010.
42
www.ijete.org

Copy-Move Forgery Detection Using DCT

Uploaded by

Copyright:

Available Formats

Copy-Move Forgery Detection Using DCT

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Copy-Move Forgery Detection Using DCT

Uploaded by

Copyright:

Available Formats

International Journal of Emerging Technologies and Engineering (IJETE)

Volume 2 Issue 3, March 2015, ISSN 2348 8050