IJERD (www.ijerd.com) International Journal of Engineering Research and Development

International Journal of Engineering Research and Development
e-ISSN : 2278-067X, p-ISSN : 2278-800X, www.ijerd.com
Volume 2, Issue 6 (August 2012), PP. 28-34

Analysis of Robustness of Digital Watermarking Techniques
under Various Attacks
Yashovardhan Kelkar, Heena Shaikh, Mohd.Imran Khan
Department of Information Technology, MIT, Ujjain

Abstract––This project acquaints an algorithm of digital watermarking which is based on Least Significant Bit (LSB),
Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT). In accord with the characters of human
vision, the main objective of the project is to be focused on comparative analysis of this algorithms on the basis of
invisibility, distortion and robustness to attacks. The simulation results show that this algorithm is invisible and has good
robustness for some common image processing operations.By the use of Matlab software, the algorithms have been coded
and then implemented properly.

Keywords––Digital watermarking, Least Significant Bit,Discrete Cosine Transform, Discrete Wavelet Transform, Peak
Signal to Noise Ratio, Mean squared Error.

I. INTRODUCTION
At the present time, the aptitude in contemplation of accessing as well as sharing images has become
progressively facile with the Internet allowing people to procure information remotely from anywhere in the entire world.
Moreover, there has been also an expansion with regard to the number of the still digital images over the internet for the sake
of the fact that a vast number of millions of people are capturing digital photos. This mentioned fact could bring forth the
requirement for people to conserve their own images or intellectual properties. Given the motivation to protect intellectual
property, Digital Watermarking technology has been referred to as fit for acceptance as a form of copyright protection and a
preventing those who have such an ambition in order to get a hold of such multimedia date either image
disproportionately[2,3].
Fundamentally, the procedure of digital watermarking can be delineated as a method for embedding information
into another signal (a digital signal). In case of digital images, the embedded information can be either visible or hidden from
the user. In this project, we will concentrate on imperceptible watermarks. The principal intention of digital watermarks is
to provide copyright protection for intellectual property that is in digital format. Typical usage scenarios for watermarking
are such as copyright protection and data authentication.Based on the digital watermarking for a still image,we are to
formulate and encode the following methods:

 The embedding and detecting procedure for watermarking technique based on LSB transform.[1][2][3]
 The embedding and detecting procedure for watermarking technique based on DCT transform.[4]
 The embedding and detecting procedure for watermarking technique based on DWT transform.[4][5]
 Computing PSNR function (peak single-to-noise ratio) the resultant watermarked images from the techniques
LSB/DCT/DWT for the purpose of measuring the distinctive distortion between the cover image and the
watermarked image.[5]
 Applying the checkmark software by means of MSE function for the original watermarks and extracted
watermarks from the LSB/DCT/DWT techniques.[5]

II. LEAST SIGNIFICANT BIT
LSB takes the watermark image embedded into the most unimportant places of vector images.This algorithm is
very simple ,strong , real time , embedded stack information and can be accurate resume embedded information. One of the
first techniques for watermarking is the least-significant bit modification. It is based on the substitution of LSB plane of the
cover image with the given watermark[3].

28

Analysis of Robustness of Digital Watermarking Techniques under Various Attacks

Fig. I Least Significant bit substitution

Make a function that replaces the least-significant bit plane of a grayscale cover image and replaces it with your
binary watermark. Since your watermark is smaller than the cover image, you need tile several copies of you watermark
together in order to achieve the correct size. You can assume that the dimensions of the cover image are divisible with the
dimensions of the watermark. The function should take as input a cover image and a watermark. Thereafter, make a function
that extracts your watermark from a watermarked image. The function should take as input a watermarked image and give
the extracted watermark as output. The functions should be called embed_lsb and extract_lsb[3,4].

III. DISCRETE COSINE TRANSFORM(DCT)
The discrete cosine transform (DCT) is a function that has the ability to convert a signal into elementary frequency
components. It represents an image as a sum of sinusoids of varying magnitudes and frequencies. The popular block-based
DCT transform segments an image non-overlapping block and applies DCT to each block. This result in giving three
frequency sub-bands: low frequency sub band, mid-frequency sub-band and high frequency sub-band. DCT-based
watermarking is based on two facts. The first fact is that most of the signal energy lies at low-frequencies sub band which
contains the most important visual parts of the image. The second fact is that high frequency components of the image
are usually removed through compression and noise attacks[5]. The watermark is therefore embedded by modifying the
coefficients of the middle frequency sub band so that the visibility of the image will not be affected and the watermark will
not be removed by compression. With an input image, x, the DCT coefficients for the transformed output image, y, are
computed according to Eq. shown below. In the equation, x, is the input imagehaving N x M pixels, x(m,n) is the intensity of
the pixel in row m and column n of the image, and y(u,v) is the DCT coefficient in row u and column v of the DCT
matrix[6].

29


The image is reconstructed by applying inverse DCT operation according to Eq. shown below:

As defined above.

1. Embedding Technique for DCT
One such a vital technique utilizes the comparison of middle-band DCT coefficients to encode a single bit into
a DCT block. There are two major processes, encoding and embedding processes[5].To commence, we first need to go
through the initial method that is the encoding and break the image into 8 x 8 blocks then apply the DCT function and
computing all the image blocks separately and converting them into frequency components based on the frequency
domain and having the whole details and components of the cover image and the watermark converted from the spatial
domain into the frequency domain.Secondly, the second procedure which is the embedding process, thus, we have to
clarify the middle-band frequency components (FM) of the image that consists of 8 x 8 DCT block. Then, generating
the reduced image which is composed of middle-band frequency components and applying it for the watermark
blocks; and with regard to the quantization table, we can select the two coefficients where we can insert the bits
needed to embed the watermark. As a result, the watermarked image will be attained[6].

Cover Image Test Image

Watermarked Image
Fig. II Embedding Technique for DCT

2. Extraction Technique for DCT
This approach (watermark detection) could be attained by a blind detection either non-blind watermark detection.
On assumption of the non-blind detection of a watermark, we have two operations, detecting and extracting process.
Through the first conduct,we need to have the cover image, the watermarked image either the watermark signature and
break them up into 8 x 8 blocks[8]. Then we should apply the DCT function and transform them into the frequency domain.
The extracting process can then start by identifying the middle frequency components (F M region) for the images and
selecting their two coefficients by means of the quantization table; so that we can generate the reduce images and
applying them to the watermark blocks and start to detect the bits where the watermark was embedded into therefore, the
watermark signature will be extracted.

30


31


Watermarked Image Retrieved Image
Fig. III Extraction Technique for DCT

IV. DISCRETE WAVELET TRANSFORM
The DWT transform, Wavelets are special functions which, in a form analogous to sines and cosines in Fourier
analysis, are used as basal functions for representing signals. For 2-D images, applying DWT corresponds to processing the
image by 2-D filters in each dimension. The filters divide the input image into four non-overlapping multi-resolution sub-
bands (LL1), (LH1), (HL1) and (HH1). The sub-band (LL1) represents the coarse-scale DWT coefficients while the sub-bands
(LH1), (HL1) and (HH1) represent the fine scale of DWT coefficients. To obtain the next coarser scale of wavelet
coefficients, the sub-band (LL1) is further processed until some final scale “N” is reached. When “N” is reached we will have
3N+1 sub-bands consisting of the multi-resolution sub-bands (LLN) and (LHX),(HLx) and (HHx) where “X” ranges from 1
until “N”. Due to its excellent spatio-frequency localization properties, the FDWT is very suitable to identify the areas in the
host image where a watermark can be embedded effectively. In particular, this property allows the exploitation of the
masking effect of the human visual system such that if a DWT coefficient is modified, only the region corresponding
to that coefficient will be modified. In general most of the image energy is concentrated at the lower frequency sub-bands
(LLX) and therefore embedding watermarks in these sub bands may degrade the image significantly. Embedding in the low
frequency sub-bands, however, could increase robustness significantly. On the other hand, the high frequency sub-bands
(HHX) include the edges and textures of the image and the human eye is not generally sensitive to changes in such sub-
bands.[9] This allows the watermark to be embedded without being perceived by the human eye. The compromise adopted
by many DWT-based watermarking algorithm, is to embed the watermark in the middle frequency sub-bands (LHX) and
(HLX) where acceptable performance of imperceptibility and robustness could be achieved.[8]

1. Encoding Technique For DWT
This technique will decompose the cover image of the two dimensional DWT into four frequency bands through the first
pass as (LL1), (LH1), (HL1) and (HH1) frequency coefficients. The frequency bands where it has the lowest resolution of
st nd
the 1 pass (LL1) can be also decomposed into a 2 level (pass). Secondly, we are to apply the Gaussian Noise and can
insert the watermark signature into the reset of the available frequency bands which include the high frequency coefficients
without dealing with (LL) regions from all over the passes (levels). We must add the signal of the bands where the large
frequency components to the signal of the Gaussian Noise and modifying them without moderating the original signal which
resides in the (LL) band; thereafter, the watermarked image would be performed appropriately.[8,9]

(a) (b)
Fig.IV(a) DWT sub bands of original message in level 2, (b) DWT sub bands of original message in level 3

Cover Image Test Image

32


Watermarked Image
Fig. V Embedding Technique for DCT

2. Decoding Technique For DWT
In contemplation of achieving this procedure, we should have the cover image and the watermarked image readily
applicable. Consequently, the DWT decoding technique will decompose those two images into four frequency bands
st
through the 1 pass as described previously [10]. Afterward, we are to select for instance one of those bands where the
large frequencies reside through one of the levels (passes) in the decomposed cover image and the decomposed
watermark. Let’s suppose the selected band from both decomposed images is (HH1), we have then to compare the difference
of the frequency coefficients in those bands of the decomposed images and examine their cross correlation.
Subsequently, if the cross correlation has detected a peak, then the watermark signature will be extracted; if not, then the
same operation will continue on comparing the rest of the other bands consist the high frequency components from both of
the decomposed images and investigate their cross correlation until the peak is detected; correspondingly, the watermark
signature will be latterly recovered[9,10].

Watermarked Image Retrieve Image
Fig. VI Extraction Technique for DCT

V. PERFORMANCE EVALUATION
This section presents the simulations and experiments of the proposed scheme and the results obtained.For
quantitative evaluation, PSNR (Peak Signal-to-Noise Ratio) is introduced to evaluate the performance of the proposed
scheme and image quality, which is defined as

Where m × n is the image size, ai,j and bi,j are the corresponding pixel values of two images.

Imperceptibility: Imperceptibility means that theperceived quality of the host image should not be distorted by the presence
of the watermark. As a measure of the quality of a watermarked image, the peak signal to noise ratio (PSNR) is typically
used.PSNR in decibels (dB).

Robustness: Robustness is a measure of the immunity of the watermark against attempts to remove or degrade it,
internationally or unintentionally, by different types of digital signal processing attacks. In this chapter, we will report on
robustness results which we obtained for three major digital signal processing operations (attacks): Gaussian noise, image
compression, salt and pepper. That is the Gaussian noise is a watermark degrading attack, JPEG compression is a watermark
removal attack.

33


Table I. Comparison of PSNR Values under Different Attacks

Table II. Comparison of MSE Values under Different Attacks

VI. CONCLUSION
In this paper, we proposed a watermarking scheme based on discrete cosine transform, wavelet packet transform
and Least significant bit. The Least Significant Bit (LSB) discrete wavelet transforms (DWT) and the discrete cosine
transform (DCT) have been applied successfully in many in digital image watermarking. We tested our proposed
watermarking scheme through extensive experiments. The experimental results clearly demonstrate the much improved
performance of the proposed approach in terms of imperceptibility and robustness against attacks.

REFERENCES
[1]. Seitz J., Digital Watermarking For Digital Media, Information Science Publishing, United States of America, May 2005.
[2]. Cox I. J., Miller M. L., Bloom J. A., Digital watermarking. San Francisco: Morgan Kaufmann Publishers, 2001.
[3]. Sanjay Rawat1, Balasubramanian Raman2. A New Robust Watermarking Scheme For Color Images, IEEE 2nd International
Advance Computing Conference 2010.
[4]. A.M.Kothari, A.C.Suthar, R.S.Gajre. Performance Analysis of Digital Image Watermarking Technique–Combined DWT–DCT
over individual DWT, Published in International Journal of Advanced Engineering & Applications, Jan. 2010.
[5]. Athanasios Nikolaidis and Ioannis Pitas. Region-Based Image Watermarking, Published in IEEE Transactions on Image
Processing, Vol. 10, No. 11, November, 2001.
[6]. I. J. Cox, M. L. Miller, J. A. Bloom, “Digital Watermarking”, Morgan Kaufmann Publishers, 2001.
[7]. Chandra Mohan B and Srinivas Kumar S. Robust Multiple Image Watermarking Scheme using Discrete Cosine Transform with
Multiple Descriptions. Published in International Journal of Computer Theory and Engineering, Vol. 1, No. 5, December, 2009.
[8]. Shital Gupta, Dr Sanjeev Jain A Robust Algorithm of Digital Image Watermarking Based on Discrete Wavelet Transform.
Published in Special Issue of IJCCT Vol.1 Issue 2, 3, 4; 2010 for International Conference [ACCTA-2010], 3-5 August 2010.
[9]. Y. Kim, Kwon, and R. Park, “Wavelet Based Watermarking Method for Digital Images Using the Human Visual System”,
Proceeding of IEEE International symposium on circuits and systems,Vol. 4, pp. 80-83. July 1999.
[10]. Lin, S. and C. Chin, “A Robust DCT-based Watermarking for Copyright Protection,” IEEE Trans.Consumer Electronics, 46(3):
415-421, 2000.

34

IJERD (www.ijerd.com) International Journal of Engineering Research and Development

More Related Content

What's hot (19)

Viewers also liked (8)

Similar to IJERD (www.ijerd.com) International Journal of Engineering Research and Development (20)

More from IJERD Editor (20)

Recently uploaded (20)

IJERD (www.ijerd.com) International Journal of Engineering Research and Development