Application of Contourlet Transform and Maximum Entropy On Digital Image Watermarking

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Journal of Theoretical and Applied Information Technology

30th September 2014. Vol. 67 No.3


2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

E-ISSN: 1817-3195

APPLICATION OF CONTOURLET TRANSFORM AND


MAXIMUM ENTROPY ON DIGITAL IMAGE
WATERMARKING
1

NADIA IDRISSI, 2AHMED ROUKHE


1,2
Faculty of Sciences Moulay Ismail Univerity Meknes, LAMPE Laboratory, Department of physics
E-mail: 1idrissi.nadia@ymail.com, 2a-roukhe@yahoo.fr

ABSTRACT
During the last few years, the copyright in multimedia world is becoming a real concern especially with the
increase in transfer files, documents over networks. These data are actually very easy to hack: information
becomes vulnerable to interception, copying, tampering or corruption. To meet this need, digital
watermarking is a method of signal processing which inserted into a digital document an invisible
watermark, containing a code robust against any attack that can affect the watermarked data. In this paper
we presented a method of digital image watermarking based on contourlet transform and maximum
entropy.
Keywords: Contourlet Transform; Entropy; PSNR; NC
1.

INTRODUCTION
2.

The network growth and increased transmission of


digital data on Internet has evolved the risk of
computer hacking. The data are transferred without
a protection that make them secure. This sometimes
prevents users to limit their data transfer. To solve
this problem, we must use techniques and methods
that enable users to protect and mark their data.
Watermarking is a method to insert a secret
message in data such as files, images and videos.
This protects the intellectual properties of these
data [1]. The insertion method should be
imperceptible to not detect the existence of a
message; it should be robust against all
conventional or geometrical attacks, as it must have
a maximum amount of information that is inserted
[2]. Watermarking methods can be applied in
various domain of digital image processing as the
spatial domain, frequency domain, and multidirectional domain. For this we will propose a
method of watermarking that present a scheme in
multi-directional domain to ensure a very strong
robustness against several attacks.
For that this paper will be organized as follows.
Section 2 illustrates the decomposition techniques
of images processing. Section 3 presents the
proposed approach for watermarking scheme in
detail. Section 4 clarifies experiment results as well
as some discussions. Conclusions are given in
section 5.

DECOMPOSITION TECHNIQUES
IMAGES PROCESSING

OF

2.1 Discrete Contourlet Transform


The Contourlet Transform (CT) is a geometrical
image- based transform [3]. Its a multi resolution
and multidirectional transformation technique that
is used in image analysis for capturing contours and
fine details in images [4]. Contourlet transform
offers multi- scale decomposition of the image; this
is obtained by using Laplacian Pyramid (LP). LP is
first used to capture point discontinuities (Figure1)
It is then followed by a directional filter bank
(DFB) to link point discontinuities into linear
structures [5] and to capture the high-frequency
content like smooth contours and directional edges.

637

Figure1. Contourlet lter Bank.

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

E-ISSN: 1817-3195

The LP decomposition at each level generates a


down sampled lowpass version of the original and
the difference between the original and the
prediction, resulting in a bandpass image. Figure 2
shows this process, where H and G are called
analysis and synthesis filters respectively and M is
the sampling matrix [6].

Figure4. Contourlet decomposition of Lena.


2.2 ENTROPY

Entropy is a statistical measure of randomness


that can be used to characterize the texture of the
input image. Entropy is defined as:
E=-sum(p.*log2(p))
Figure2. Laplacian Pyramid Scheme (A) Analysis And
(B) Reconstruction.

The bandpass image obtained in the LP


decomposition is further processed by a DFB. A
DFB is designed to capture the high-frequency
content like smooth contours and directional edges
[7, 8]. The DFB is efficiently implemented via a Klevel binary tree decomposition that leads to 2K
subbands
with
wedge-shaped
frequency
partitioning as shown in Figure 3. The contourlet
decomposition is illustrated by using the Lena test
image of size 512512 and its decomposition into
four levels, in Figure 4. At each successive level,
the number of directional subbands is 2, 4, 8, and
16 [8].

Where p contains the histogram counts returned


from imhist. By default, entropy uses two bins for
logical arrays and 256 bins for uint8, uint16, or
double arrays. [9-11].
3.

PROPOSED METHOD

In this paper, contourlet and SVD based


transform technique is proposed for watermarking
of gray scale image along with binary watermark
logo. The robustness and imperceptibility of
watermarked image is increased by selecting
specifics blocks which have the maximum entropy
value. And is tested for two quantifiers such as
PSNR.
3.1 Watermark embedding process
The embedding process is divided into
following steps and is briefly described as bellow:
Step 1: Apply contourlet transform on the original
image; Apply contourlet transform on the
watermark.
Step 2: extration of coefficients with different
orientations.
Step 3: division of each orientation in blocks of
same sizes,

Figure3. Laplacian Pyramid Scheme (A) Analysis And


(B) Reconstruction.

Step 4: modify the original image with watermark


such that:

S = I + W
Step 5: Apply inverse Contourlet Transform

=0, /2, /4 and 3/4

Step 7: obtain the watermarked image.


638

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

This scheme resumed the steps of embedding


process

Original
image I

Watermark
W
Step 1

E-ISSN: 1817-3195

The robustness is tested under 9 types of


attacks: Salt and pepper, Gaussian noise, median
filtering, resizing, contrast adjustment, JPEG
compression, cropping, and rotation, Histogram
equalization.
The original image considered for the
experimentation is Lena. The watermark image is a
logo as shown in Figure 6. Figure7 indicates that
visual appearance of the watermarked image is
good with a PSNR of 60.5446 Db.

Step 2

W1

I1

I2

(a)

W2

(b)

Step 3
I3

I4

W3

Figure6. (a) Original watermark, (b) extracted


watermark

W4

Step 4

S = I + W

(a)

(b)

Figure7. (a) Original image, (b) watermarked image

=0, /2, /4 and 3/4


Step 5

To improve the performance of our proposed


method we have used other different standard
images test.

Apply inverse Contourlet


Step 6
Camerama

Watermarked
Image
Figure5. Watermark Embedding
Process

Jetplane

Figure8. Images of test

The metric used to assess the performance


of the proposed algorithm is PSNR.
The results of different attacks of lena image are
presented in figures as bellow:

4.

EXPERIMENTAL RESULTS
639

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

E-ISSN: 1817-3195

Resizing

(a)

(b)
(a)
PSNR=20.2872dB

PSNR= 11.1127dB
Rotation (200)

(a)

Contrast adjustement

(b)

(a)
PSNR=29.2389dB

PSNR= 18.7821dB

(b)

JPEG QF=70

Gaussian noise

(a)

(b)

(a)
PSNR=5.2673 dB

(b)

(b)

Histogram equalization

PSNR= 25.0552dB

Figure 9(a).Attacked watermarked image with PSNR.


(b).Extracted watermark from LH matrix.

Median filtering (3x3) kernel

These results are resumed in table and graphs as

640

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

E-ISSN: 1817-3195

bellow:
TABLE I.

PSNR Results After Various Attacks For


Lena Image
Attacks

PSNR

Without attack

60.5446

Salt and pepper noise (0.01


density)

25.1734

Rotation (200)

11.1127

Gaussian noise

18.7821

Median filtering (3x3) kernel

25.0552

Resizing

12.3331

Contrast adjustement

20.2872

Cropping 25%

12.3548

Histogram equalization

5.2673

JPEG QF=70

29.2389

TABLE II.

Psnr Results After Various Attacks For


Cameraman Image
Attacks

PSNR

Without attack

61.5553

Salt and pepper noise (0.01


density)

25.2665

Rotation (200)

11.1127

Gaussian noise

17.2357

Median filtering (3x3) kernel

25.1372

Resizing

12.3331

Contrast adjustement

20.2872

Cropping 25%

11.0325

Histogram equalization

5.2673

JPEG QF=70

29.9965

Attacks

PSNR

Without attack

59.3855

Salt and pepper noise (0.01


density)

24.7875

Rotation (200)

11.1127

Gaussian noise

18.3985

Median filtering (3x3) kernel

24.7139

Resizing

12.3331

Contrast adjustement

20.2872

Cropping 25%

12.1398

Histogram equalization

5.2673

JPEG QF=70

29.1782

Figures 10 and 11 shows the comparison between


the variation of the PSNR value without attacks and
after application of various attacks on several test
images. These figures demonstrate that cameraman
image has the best PSNR before applying
attacksand jetplane image has the best PSNR after
applying all attacks used in our approach.

Figure10. PSNR Variations For Different Image Without

Applying Attacks

TABLE III.

PSNR Results After Various Attacks For


Jetplane Image

641

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

Figure11. PSNR Variations For Different Image With Applying Attacks

5.

CONCLUSION

In this paper, we proposed a new approach for


watermarking of still images based on both
contourlet
transform
and
singular
value
decomposition concepts. The experimental results
show that this method is robust against most attacks.
Thereby we can conclude that intellectual property
of images is protected by our approach. Future work
will focus on a more robust method that combines
its characteristics with those of contourlet transform
for the best results. This approach can also be
extended to video and audio watermarking.
REFRENCES:
[1] Natarajan Meghanathan, Lopamudra Nayak,
Steganalysis algorithms for detecting the
hidden information in image, audio and video
cover media, International Journal of Network
Security & Its Application (IJNSA), Vol.2,
No.1, January 2010.
[2] Mr.Manjunatha Prasad.R, Dr.Shivaprakash
Koliwad, A Comprehensive Survey of
Contemporary Researches in Watermarking for
Copyright Protection of Digital Images,
IJCSNS International Journal of Computer
Science and Network Security, VOL.9 No.4,
April 2009.
[3] M. N. Do and M. Vetterli, The contourlet
transform:
An
ecient
directional
multiresolution image representation, IEEE
Transactions on Image Processing, vol. 14, no.
12, pp.20912106, 2005.

E-ISSN: 1817-3195

[4] C.Venkata Narasimhulu & K.Satya Prasad, A


hybrid watermarking scheme using contourlet
transform and singular value decomposition ,
IJCSNS International Journal of Computer
Science and Network Security, VOL.10 No.9,
September 2010.
[5] Minh N. Do, Member, IEEE, and Martin
Vetterli,Fellow, IEEE, The Contourlet
Transform:
An
Efficient
Directional
Multiresolution Image Representation, IEEE
TRANSACTIONS
ON
IMAGE
PROCESSING, Volume: 14 , Issue: 12 2005.
[6] Nagham S. AL-lella, Khalil I. Alsaif, Data
hiding in contourlet coefficients based on their
energy , J. of university of anbar for pure
science : Vol.6:NO.2 : 2012.
[7] Ibrahim A. El rube, Mohamad Abou El Nasr ,
Mostafa M. Naim, Mahmoud Farouk,
Contourlet versus wavelet transform for a
robust digital image watermarking technique,
World academy of science, Engeneering and
technology Vol:36 2009-12-21.
[8] Sirvan Khalighi, Parisa Tirdad and HamidR
Rabiee, A Contourlet Based Image
Watermarking Scheme With High Resistance
to Removal and Geometrical Attacks,
EURASIP Journal on Advances in Signal
Processing 2010, 2010:540723.
[9] Mohamed RADOUANE, Tarik BOUJIHA,
Rochdi MESSOUSSI, Nadia IDRISSI, Ahmed
ROUKH, A Method of LSB substitution
based on image blocks and maximum entropy,
IJCSI International Journal of Computer
Science Issues, Vol. 10, Issue 1, No 1, January
2013.

642

Journal of Theoretical and Applied Information Technology


30th September 2014. Vol. 67 No.3
2005 - 2014 JATIT & LLS. All rights reserved.

ISSN: 1992-8645

www.jatit.org

[10] Nadia IDRISSI, Ahmed ROUKH, Lhoussaine


MASMOUDI,
Mohamed
RADOUANE,
Rochdi MESSOUSSI, A Robust Digital
Watermarking Technique using DWT-DCT
and Statics blocks, IJCSI International Journal
of Computer Science Issues, Vol. 10, Issue 2,
No 1, March 2013.
[11] Mohamed RADOUANE, Tarik BOUJIHA,
Rochdi MESSOUSSI, Raja TOUAHNI, A
ROBUST METHOD FOR DIGITAL IMAGES
WATERMARKING
BASED
ON
COMBINATION OF SVD, DWT AND DCT
USING OPTIMAL BLOCK,JATIT, VOL59,
No2 20th January 2014.

643

E-ISSN: 1817-3195

You might also like