SECURED COLOR IMAGE WATERMARKING TECHNIQUE IN DWT-DCT DOMAIN

International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol.1, No.3, August 2011
DOI : 10.5121/ijcseit.2011.1304 36
SECURED COLOR IMAGE WATERMARKING
TECHNIQUE IN DWT-DCT DOMAIN
Baisa L. Gunjal1
and Suresh N.Mali2
1
Amrutvahini College of Engineering, Sangamner, A’nagar, MS, India
hello_baisa@yahoo.com
2
Imperial College of Engineering and Research, Wagholi, Pune, MS, India
snmali@rediffmail.com
ABSTRACT
The multilayer secured DWT-DCT and YIQ color space based image watermarking technique with
robustness and better correlation is presented here. The security levels are increased by using multiple pn
sequences, Arnold scrambling, DWT domain, DCT domain and color space conversions. Peak signal to
noise ratio and Normalized correlations are used as measurement metrics. The 512x512 sized color images
with different histograms are used for testing and watermark of size 64x64 is embedded in HL region of
DWT and 4x4 DCT is used. ‘Haar’ wavelet is used for decomposition and direct flexing factor is used. We
got PSNR value is 63.9988 for flexing factor k=1 for Lena image and the maximum NC 0.9781 for flexing
factor k=4 in Q color space. The comparative performance in Y, I and Q color space is presented. The
technique is robust for different attacks like scaling, compression, rotation etc.
KEYWORDS
DCT-DWT, Scrambling, Histogram, YIQ color space.
1. INTRODUCTION
It has become a daily need to create copy, transmit and distribute digital data as a part of
widespread multimedia technology in internet era. Hence copyright protection has become
essential to avoid unauthorized replication problem. Digital image watermarking provides
copyright protection to image by hiding appropriate information in original image to declare
rightful ownership [1]. Robustness, Perceptual transparency, capacity and Blind watermarking are
four essential factors to determine quality of watermarking scheme [4][5]. Watermarking
algorithms are broadly categorized as Spatial Domain Watermarking and Transformed domain
watermarking. In spatial domain, watermark is embedded by directly modifying pixel values of
cover image. Least Significant Bit insertion is example of spatial domain watermarking. In
Transform domain, watermark is inserted into transformed coefficients of image giving more
information hiding capacity and more robustness against watermarking attacks because
information can be spread out to entire image [1]. Watermarking using Discrete Wavelet
Transform, Discrete Cosine Transform, CDMA based Spread Spectrum Watermarking are
examples of Transform Domain Watermarking. The rest of the paper is organized as follows:
Section 2 focuses on survey of color image watermarking algorithms. Section 3 gives details of
fundamentals of elements used in proposed methodology. In section 4, proposed methodology is
explained. Section 5 shows Experimental results after implementation and Testing. In section 6,
conclusion is drawn.

37
2. SURVEY
Though Fourier transform, short time Fourier transform and continuous wavelet transform are
available in transform domain, but all of them are having their own limitations. Discrete Wavelet
Transform provides multi resolution for given image and can efficiently implemented using
digital filters, it has become attraction of researchers in image processing area. Here, review of
literature survey is done on different transform in transform domain and existing color image
watermarking techniques with based on ‘Discrete Wavelet Transform. Following are some
existing methods for in color image watermarking:‘ In [10], Integer Wavelet Transform with Bit
Plane complexity Segmentation is used with more data hiding capacity. This method used RGB
color space for watermark embedding. In [2] DWT based watermarking algorithm of color
images is proposed. The RGB color space is converted into YIQ color space and watermark is
embedded in Y and Q components. This method gives correlation up to 0.91 in JPEG
Compression attack. In [3], Watermarking Algorithm Based on Wavelet and Cosine Transform
for Color Image is proposed. A binary image as watermark is embedded into green or blue
component of color image. In [4], Color Image Watermarking algorithm based on DWT-SVD is
proposed. The scrambling watermark is embedded into green component of color image based on
DWT-SVD. The scheme is robust and giving PSNR up to 42, 82. In [5], Pyramid Wavelet
Watermarking Technique for Digital Color Images is proposed. This algorithm gives better
security and better correlation in Noise and compression attacks.
3. FOUNDATIONS OF OUR METHODOLOGY
3.1. RGB and YIQ Color Space
RGB color space can be converted into YIQ color space. Y’ is similar to perceived luminance; ‘I
and Q’ carry color information and some luminance information. Since pixel values are highly
correlated in RGB color spaces, the watermark embedding in YIQ color space is preferred for
Watermarking. Initially color image is read and R, G, B components of original Cover Image are
separated. Then they are converted into YIQ color Space using following equations [2]. After
conversion of RGB color spaces into YIQ color spaces, Watermark is embedded.
Y = 0.299 ∗ R + 0.587 ∗ G + 0.114 ∗ B (1)
I = 0.596 ∗ R − 0.274 ∗ G − 0.322 ∗ B (2)
Q = 0.211 ∗ R − 0.522 ∗ G + 0.311 ∗ B (3)
After embedding the watermark using DWT, YIQ color space is converted back into RGB color
space using following equations.
R = Y + 0.956 ∗ I + 0.621 ∗ Q (4)
G = Y − 0.272 ∗ I − 0.647 ∗ Q (5)
B = Y − 1.106 ∗ I + 1.702 ∗ Q (6)
3.2. Selection of sub band in DWT
ISO has developed and generalized still image compression standard JPEG2000 which substitutes
DWT for DCT. DWT offers mutiresolution representation of image and DWT gives perfect
reconstruction of decomposed image. Discrete wavelet can be represented as
߰௝,௞ሺ‫ݐ‬ሻ = ܽ଴
ି௝/ଶ
߰ ሺܽ଴
ି௝
‫ݐ‬ − ݇ ܾ଴ሻ (7)
For dyadic wavelets a0 =2 and b0 =1, Hence we have,
߰௝,௞ሺ‫ݐ‬ሻ = 2ି௝/ଶ
߰ ሺ2ି௝
‫ݐ‬ − ݇ ሻ j, k ߳ ܼ (8)
When image is passed through series of low pass and high pass filters, DWT decomposes the
image into sub bands of different resolutions [6][7][8][9].

38
Figure 1: One Level Image Decomposition
As shown in figure 1, DWT decomposes image into four non overlapping multi resolution sub
bands: LL1 (Approximate sub band), HL1 (Horizontal sub band), LH1 (Vertical sub band) and
HH1 (Diagonal Sub band). Here, LL1 is low frequency component whereas HL1, LH1 and HH1
are high frequency (detail) components [2].To obtain next coarser scale of wavelet coefficients
after level 1, the sub band LL1 is further decomposed as per requirement. Embedding watermark
in low frequency coefficients can increase robustness significantly but maximum energy of most
of the natural images is concentrated in approximate (LL1) sub band. Hence modification in this
low frequency sub band will cause severe and unacceptable image degradation. Hence watermark
is not embedded in LL1 sub band. The good areas for watermark embedding are high frequency
sub bands (HL1, LH1 and HH1), because human naked eyes are not sensitive to these sub bands.
They yield effective watermarking without being perceived by human eyes. But HH1 sub band
includes edges and textures of the image. Hence HH1 is also excluded. The rest options are HL1
and LH1. But Human Visual System is less sensitive in horizontal than vertical. Hence
Watermarking done in HL1 region.
3.3. Selection of Block size in Discrete Cosine Transform
The discrete cosine transform (DCT) represents an image as a sum of sinusoids of varying
magnitudes and frequencies. The DCT has special property that most of the visually significant
information of the image is concentrated in just a few coefficients of the DCT [3]. It’s referred as
‘Energy compaction Property’. The DCT for image A with M x N size is given by:
‫ܶܥܦ‬௣௤ = ߙ௣ߙ௤ ∑ ∑ ‫ܣ‬௠௡
ேିଵ
௡ୀ଴
ெିଵ
௠ୀ଴ cos ቀ
గሺଶ௠ାଵሻ௣
ଶெ
ቁ cos ቀ
గሺଶ௡ାଵሻ௤
ଶே
ቁ (9)
where,
0≤ ‫݌‬ ≤ ‫ܯ‬ − 1 , and 0≤ ‫ݍ‬ ≤ ܰ − 1 (10)
ߙ௣ = ቊ
1 √‫ܯ‬⁄ , ‫݌‬ = 0
ඥ2/‫,ܯ‬ 1 ≤ ‫݌‬ ≤ ‫ܯ‬ − 1
(11)
ߙ௤ = ቊ
1 √ܰ⁄ , ‫ݍ‬ = 0
ඥ2/ܰ, 1 ≤ ‫ݍ‬ ≤ ܰ − 1
(12)
As DCT is having good energy compaction property, many DCT based Digital image
watermarking algorithms are developed. It’s already proved that DWT-DCT combined approach
can significantly improve PSNR with compared to only DCT based watermarking methods.
3.4. Scrambling Method and Arnold Periodicity
Different methods can be used for image scrambling such as Fass Curve, Gray Code, Arnold
Transform, Magic square etc. Here Arnold Transform is used. The special property of Arnold
Transform is that image comes to it’s original state after certain number of iterations. These’
number of iterations’ are called ‘Arnold Period’ or ‘Periodicity of Arnold Transform’. The Arnold
Transform of image is given by
LL1 HL1
LH1 HH1

39
ቀ௫೙
௬೙
ቁ = ቂ
1 1
1 2
ቃ ቀ௫
௬
ቁ ሺ݉‫݀݋‬ ܰሻ (13)
Where, (x, y) ={0,1,.....N} are pixel coordinates from original image. (‫ݔ‬௡,‫ݕ‬௡ ) are corresponding
results after Arnold Transform. The periodicity of Arnold Transform (P), is dependent on size of
given image. From equation: 3 we have,
‫ݔ‬௡=x+y (14)
‫ݕ‬௡=x+2*y
If (mod (‫ݔ‬௡, N) ==1 && mod (‫ݕ‬௡, N) ==1) (15)
then P=N (16)
4. PROPOSED METHODOLOGY
The Watermark Embedding and Extraction Process for HL sub band for I component is
given below. Same procedure is followed applied for Y and Q components for results.
4.1 Watermark Embedding Algorithm
Step 1: Read Color Cover Image of 512x512 size. Separate it’s R,G,B components and convert
into YIQ color space using equations1,2,3.
Step 2: Now select I component and apply one level DWT. Consider HL1 sub band.
Step 3. Read grey scale watermark of 64x64 size.
Step 4: Depending upon Key K1, generate pn sequence for given watermark and calculate sum
say SUM, which is summation of all elements in generated pn sequence.
Step 5. Determine Arnold Periodicity P for given watermark.
Step 6: If SUM > T, where T is some predefined threshold value, then perform watermark
scrambling by Key K2= P+ Count, Otherwise perform watermark scrambling by Key K3= P+
Count, where count is predefined value used as counter. Here, we get ‘Scrambled Watermark’ by
Arnold Transform.
Step 7: Generate two pn sequences: pn_sequence_0 and pn_sequence_1, depending upon sum of
all elements of mid band used for 4x4 DCT transformation.
Step 8: Perform watermark embedding using following equations:
If Watermark bit is 0, then
‫ܦ‬′
= ‫ܦ‬ + ‫ܭ‬ ∗ ‫0_݁ܿ݊݁ݑݍ݁ݏ_݊݌‬ (18)
If Watermark bit is 1, then
‫ܦ‬′
= ‫ܦ‬ + ‫ܭ‬ ∗ ‫1_݁ܿ݊݁ݑݍ݁ݏ_݊݌‬ (19)
Where D is matrix of mid band coefficients of DCT Transformed block and ‫ܦ‬′
is Watermarked
DCT block.
Step 9: Apply Inverse DCT to get ‘New_HL1’ component.

40
Step 10: Apply inverse DWT with ‘LL1, New_HL1,LH1, HH1’ to get ‘New_I’ component.
Step 11: Combine, Y, New_I and Q components and convert to RGB color space using equation
4, 5,6.
K1
Figure 2: Block diagram for Watermark Embedding Process
4.2 Watermark Extraction Algorithm
Step 1: Read Color ‘Watermarked_Image’ and separate it’s R,G and B components. Now
convert to YIQ color space.
Step 2: Now select I component and apply one level DWT to retrieve HL1 sub band.
Step 3: Use 4x4 size for DCT blocks. Generate two pn sequences: pn_sequence_0 and
pn_sequence_1, depending upon sum of all elements of mid band used for 4x4 DCT
transformation. Use same seed which was used in watermark embedding process. e.g. if rand
(‘state’, 15) is used in embedding process, then, same process is to repeated here.
Step 4: Extract mid band elements from DCT block and find correlation between ‘extracted mid
band coefficients and pn_sequence_0’ as well as ‘extracted mid band coefficients and
pn_sequence_1’.
Step 5 Determine watermark bits as follows:
If correlation between ‘extracted mid band coefficients and pn_sequence_0’ is greater than
‘extracted mid band coefficients and pn_sequence_1’, then record watermark bit as 0 else record
watermark bit as 1. Here we get ‘Intermediate watermark’.
Step 6: Apply Arnold Scrambling to Intermediate watermark’ to give final recovered watermark.
No
Yes
I
Component
DWT:
HL band
Watermark PN
Sequence
Inverse
DCT
Inverse DWT
DCT4x4
DCT
Embedding
Algorithm
If
SUM>T
Scrambled by
k2=p + Count
R, G, B to
Y,U,V color
space
Original
Color Cover
Image
YIQ to RGB
Conversion
Scrambled by
k3 = p-Count
Watermarked
Image

41
Figure 3: Block diagram for Watermark Extraction Process
5. EXPERIMENTATION
Image histograms are used as important feature during the testing of this method. Image
histogram describes the distribution of pixel values into equal-sized bins. Then the range of pixels
of image falling into each bin is calculated. The style of histogram may be described by:
‫ܪ‬ = ሼℎሺ݅ሻ|݅ = 1,2, … 256|} (20)
where H is a vector denoting the volume-level histogram of intensity signal ‫ܨ‬ = ሼ݂ሺ݅ሻ|݅ =
1,2 … . . ܰ|} and ℎሺ݅ሻ, ℎሺ݅ሻ ≥ 0 denotes number of samples in bin i and satisfies ∑ ℎሺ݅ሻ = ܰே
௜ୀଵ
Different images have different pixel distribution. Hence they have different histogram shapes.
The images having difference in their histogram shapes reflect better effect of algorithm. In figure
4, five different cover images of 512x512 sizes with their R, G, B histograms are shown. They are
used as ‘Test Cover Images’ in our experiment. The watermark is of 64x64 size. The result is
tested for flexing factor k=1, 2, 3,4. The sample results are presented only for k=1 and k=4 as
shown in Table 1 and Table 2. For Lena image maximum PSNR value is 63.9988 for flexing
factor k=1. The maximum NC 0.9781 for flexing factor k=4. The PSNR and NC values for Q
channel are better than PSNR and NC values for Y and I channels. The experimentation is done
in Matlab and PSNR (Peak Signal to Noise Ratio) and NC (Normalized Correlation) are used are
measurement metrics. PSNR measures perceptual transparency’ and given by:
ܴܲܵܰሺܾ݀ሻ = 10݈‫݃݋‬ଵ଴
ሺெ௔௫಺ሻమ
భ
ಾ∗ಿ
∑ ∑ [௙ሺ௜,௝ሻି௙′ሺ௜,௝ሻ]మಿ
ೕసభ
ಾ
೔సభ
(21)
Where, f (i, j) is pixel of original image. f ‘(i, j) is pixel values of watermarked image. MaxI is the
maximum pixel value of image. NC measures robustness and given by:
ܰ‫ܥ‬ =
∑ ௪೔ ௪೔ ′ಿ
೔సభ
ට∑ ௪೔
ಿ
೔సభ ට∑ ௪೔
ಿ
೔సభ ′
(22)
where, N is total pixels in watermark, wi is original watermark, wi’ is extracted watermark.
Watermarked
Image
RGB component
Separation
RGB to YIQ
conversion
DWT:
HL band
Generation of
two pn sequences
Extraction of mid
band elements of
DCT
Correlation:: mid
band elements
& pn sequences
Scrambled
Watermark
I
Component
Final
Watermark

42
Figure 4: Images used as ‘cover images’ with histograms of Red Green and Blue Plane
Figure 5: Cover Image ‘peppers’ and Y, I, Q components and Watermarked image, and watermark
Table 1: Results of HL3 sub band for Flexing Factor k=1
Cover
Image
Histo-gram
of Red
Plane
Histo-gram
of Green
Plane
Histo-gram
of Blue
Plane
Original
Extracted
a) Cover
Image
b) Y
Component
b) I
Component
b) Q
Component
e) Watermark-
ed Image
Watermark
Cover
Image
Metric PSNR NC PSNR NC PSNR NC PSNR NC PSNR NC
Y 55.4919 0.3328 55.3712 0.3306 55.2730 0.2404 55.347 0.314 55.399 0.1278
I 55.9477 0.8111 55.8494 0.9266 55.6905 0.7077 55.450 0.903 55.816 0.7312
Q 63.1887 0.7425 63.9988 0.9355 64.6283 0.7342 67.568 0.916 63.947 0.7788

43
Table 2: Results of HL3 sub band for Flexing Factor k=4
6. CONCLUSION
In this paper a strongly robust and multilayer security based color image watermarking algorithm
in DWT-DCT domain is presented. Since pixel values are highly correlated in RGB color spaces,
the use of YIQ color space for watermark embedding is beneficial for improvement in results.
The images having difference in their histogram shapes reflect better effect of algorithm. Hence
images with different histograms are used in experiment. The PSNR and NC values for Q channel
are better than PSNR and NC values for Y and I channels. For Lena image the PSNR value is
63.9988 for flexing factor k=1. The maximum NC 0.9781 for flexing factor k=4. The average
performance of PSNR and NC for Y and I channels are approximately remains in same range for
given flexing factor. The technique is robust for different attacks like scaling, compression,
rotation etc. This algorithm provides multilayer security by using pn sequence, Arnold
scrambling.DWT domain, DCT domain, and color space conversions.
ACKNOWLEDGMENT
Thanks to BCUD, Pune for providing ‘Research Grant’ for this work. File Ref. No.-
BCUD/OSD/390 Dated 25/10/2010. We are thankful to ‘Amrutvahini College of Engineering,
Sangamner, A’nagar’ and ‘Imperial College of Engineering and Research’, Wagholi, MS, India
for providing technical support during the work.
REFERENCES
[1] Cheng-qun Yin et al, “ Color Image Watermarking Algorithm Based on DWT-SVD”, Proceeding of
the IEEE International Conference on Automation and Logistiocs August 18-21, 2007, Jinan, China,
PP: 2607-2611
[2] Guangmin Sun, Yao Yu, “ DWT Based Watermarking Algorithm of Color Images”, Second IEEE
Conference on Industrial Electronics and Application”,2007, PP 1823-1826.
[3] Wei-Min Yang, Zheng Jin, “ A Watermarking Algorithm Based on Wavelet and Cosine Transform for
Color Image”, First International Workshop on Education Technology and Computer
Science”, 2009, PP 899,903.
[4] Cheng-qun Yin, Li Li, An-qiang Lv and Li Qu, “ Color Image Watermarking Algorithm Based on
DWT-SVD”, Proceeding of the IEEE International Conference on Automation and Logistics , August
18-21, 2007, Jinan, China, PP 2607-2611.
Cover
Image
Metric PSNR NC PSNR NC PSNR NC PSNR NC PSNR NC
Y 44.7861 0.7157 44.8162 0.7669 44.8008 0.5035 44.78 0.613 44.8000 0.3477
I 45.1319 0.9763 45.0086 0.9781 44.9074 0.9746 44.92 0.978 44.9621 0.9727
Q 48.6263 0.9781 48.5319 0.9781 48.4816 0.9781 48.46 0.978 48.5158 0.9781

International Journal of Computer Science, Engineering
[5] Awad Kh. Al-smari and Farhan A. Al
Copyright Protection”, International Conference on Computing,
44-47
[6] B.L.Gunjal, R.R.Manthalkar, “Discrete Wavelet Transform Based Strongly Robust Watermarking
Scheme for Information Hiding in Digital Images”, Third International Conference
in Engineering and Technology,19
http://doi.ieeecomputersociety.org/10.1109/ICETET.2010.12
[7] S. Joo, Y. Suh, J. Shin, H. Kikuchi, and S. J. Cho., “A new robust watermark embedding into wavelet
DC components,�� ETRI Journal,
[8] Voloshynovskiy. S. S. Pereira and T. Pun. 2001. “Attacks on Digital watermarks: classification,
Estimation-Based attacks and Benchmarks”, Comm, Magazine. 39(8):118
[9] Abu-Errub, A., Al-Haj, A.,”Optimized DWT
Conference on Applications of Digital Information and Web Technologies, IEEE,2008, 4
[10] K.Ramani, E Prasad, S Varadarajan, “Stenography using BPCS to the integer wavelet
transform”,IJCSNS international journal of Comput
2007.
Authors
B.L.Gunjal completed her B.E. Computer from University of Pune and M.Tech in
I.T. in Bharati Vidyapeeth, Pune , Maharashtra, india. She is having 13 Years
teaching experience and 18 international and national publications. Presently she is
working on research project on “Image Watermarking” funded by BCUD,
University of Pune. Her areas of interest includes Image Processing, Advanced
databases and Computer Networking.
Dr. Suresh N. Mali has completed his PhD form Bharati Vidyapeeth Pune, presently
he is working as Principal in Imperial College of Engineering and Research
,Wagholi, Pune. He is author of 3 books and having more than 18 international and
national publications. He is working
New Delhi and also working as BOS member for Computer Engineering at
University of Pune. He has worked as a Member of Local Inquiry Committee to
visit at various institutes on behalf of University of Pune. His areas
mainly includes Image Processing
smari and Farhan A. Al-Enizi, “A Pyramid-Based Technique for Digital Color Images
Copyright Protection”, International Conference on Computing, Engineering and Information, 2009, pp
B.L.Gunjal, R.R.Manthalkar, “Discrete Wavelet Transform Based Strongly Robust Watermarking
Scheme for Information Hiding in Digital Images”, Third International Conference- Emerging Trends
echnology,19-21 Nov 2010 , Goa, India, ISBN 978-0-7695-4246-1,
http://doi.ieeecomputersociety.org/10.1109/ICETET.2010.12.
S. Joo, Y. Suh, J. Shin, H. Kikuchi, and S. J. Cho., “A new robust watermark embedding into wavelet
DC components,” ETRI Journal, 24, 2002, pp. 401-404.
Voloshynovskiy. S. S. Pereira and T. Pun. 2001. “Attacks on Digital watermarks: classification,
Based attacks and Benchmarks”, Comm, Magazine. 39(8):118-126.
Haj, A.,”Optimized DWT-based image watermarking”, First International
Conference on Applications of Digital Information and Web Technologies, IEEE,2008, 4
K.Ramani, E Prasad, S Varadarajan, “Stenography using BPCS to the integer wavelet
transform”,IJCSNS international journal of Computer science and network security, vol
B.L.Gunjal completed her B.E. Computer from University of Pune and M.Tech in
I.T. in Bharati Vidyapeeth, Pune , Maharashtra, india. She is having 13 Years
ational and national publications. Presently she is
working on research project on “Image Watermarking” funded by BCUD,
University of Pune. Her areas of interest includes Image Processing, Advanced
databases and Computer Networking.
has completed his PhD form Bharati Vidyapeeth Pune, presently
he is working as Principal in Imperial College of Engineering and Research
,Wagholi, Pune. He is author of 3 books and having more than 18 international and
national publications. He is working as Member of expert Committee of AICTE,
New Delhi and also working as BOS member for Computer Engineering at
University of Pune. He has worked as a Member of Local Inquiry Committee to
visit at various institutes on behalf of University of Pune. His areas of interest
and Information Technology (IJCSEIT), Vol.1, No.3, August 2011
44
Based Technique for Digital Color Images
Engineering and Information, 2009, pp
B.L.Gunjal, R.R.Manthalkar, “Discrete Wavelet Transform Based Strongly Robust Watermarking
Emerging Trends
1,
S. Joo, Y. Suh, J. Shin, H. Kikuchi, and S. J. Cho., “A new robust watermark embedding into wavelet
Voloshynovskiy. S. S. Pereira and T. Pun. 2001. “Attacks on Digital watermarks: classification,
watermarking”, First International
Conference on Applications of Digital Information and Web Technologies, IEEE,2008, 4-6.
K.Ramani, E Prasad, S Varadarajan, “Stenography using BPCS to the integer wavelet
er science and network security, vol-7,No: 7 July

SECURED COLOR IMAGE WATERMARKING TECHNIQUE IN DWT-DCT DOMAIN

Related slideshows

More Related Content

SECURED COLOR IMAGE WATERMARKING TECHNIQUE IN DWT-DCT DOMAIN