Color image multiple watermarking scheme based on discrete wavelet transform

Color image multiple watermarking scheme based on discrete wavelet transform

Proceedings of International Conference ICSEM13

N. MOHANANTHINI

Department of Electrical Engineering Annamalai University

Tamilnadu, India

e-mail: mohananthini@yahoo.co.in

G. YAMUNA

Department of Electrical Engineering Annamalai University

Tamilnadu, India

1. ail: yamuna.sky@gmail.com

   Abstract – Digital Watermarking is an important technique in the area of information security. The proposed scheme is based on Discrete Wavelet Transform (DWT). The embedding and extraction process using multi-resolution analysis of wavelet Transform. A developed algorithm embeds the watermark, information without much distortion to the image and to extract the watermark, information in the absence of original image. The proposed method has good imperceptibility on the watermarked image, superior in terms of Peak Signal to Noise Ratio (PSNR).

   Keywords – Correlation; Discrete Wavelet Transform; Multiple watermarking; PSNR.

   1. INTRODUCTION

      Digital watermarking technology has evolved very quickly these years. Digital watermarking is an important specialized area for the advancement of image processing technology. The image watermarking technique is one of the important techniques which are used for safeguarding the origins of the image by protecting it against Piracy. Digital watermarking is the process of embedding information into a digital image in a way that is difficult to remove.

      Watermarking techniques can be broadly classified into two categories: spatial domain methods and transform domain methods. For transform domain schemes, the host image is first converted into frequency coefficients by a transformation method such as the discrete cosine transforms (DCT), discrete Fourier transforms (DFT) and discrete wavelet transforms (DWT). Among the transform domain watermarking techniques, wavelet transform based watermarking techniques are gaining more popularity. Numerable work relating to DWT watermarking method has been reported.

      A detail survey on wavelet based watermarking techniques can be found in [1]. To elaborates suitability of wavelet transform for image watermarking, wavelet transform based image watermarking process, classification and analysis of wavelet based watermarking techniques is proposed in [2]. Digital watermarking on still images using Wavelet transform is proposed in [3]. Xia et al. [4] has proposed twolevel decomposition using the Haar wavelet filters. Their method the pseudo Random Noise codes are only added to the large coefficients of the

      high and middle frequency bands of the DWT transformed image. An image accreditation technique based on DWT by embedding digital watermarks in images is proposed in [5].

      A novel image watermarking technique in the wavelet domain is suggested and tested in [6]. Their method to achieve more security and robustness, their techniques relies on using two nested watermarks that are embedded into the image to be watermarked. A primary watermark in form of a PN sequence is first embedded into an image (the secondary watermark) before being embedded into the host image. Peter et al. [7] has proposed three novel blind watermarking schemes to embed watermarks into digital images. The single watermark embedding (SWE) can embed one watermark bit sequence. The multiple watermark embedding (MWE) can use correlated keys to embed multiple watermark bit sequences simultaneously such that individual watermark bit sequence can be decoded or detected independently. The iterative watermark embedding (IWE) can embed watermark in a JPEG file and ensure it is detectable. Experimental results shows that the three proposed watermarking algorithms give watermarked images with good visual quality. Multiple watermarks can be used to address multiple applications or one application may be addressed several times [8]. Two visual watermarks are embedded in the DWT domain through modification of both low and high frequency coefficients are explained in [9]. A novel robust multiple watermarking techniques for color images in spatial domain are proposed in [10]. Their method the host image is divided into four different regions. An invisible watermarking technique is proposed in [11], to embed multiple binary watermarks into digital images based on the concept of Visual Cryptography (VC).

      The performance of orthogonal and biorthogonal wavelet filters for image compression presented in [12]. They evaluated their method by objectively and subjectively. A new color image watermarking scheme based on the color quantization technique is proposed in [13]. Experimental results are shown to demonstrate the validity of the proposed scheme, which can be applied to other multimedia applications that are based on the color quantization technique. A grayscale visual watermark image is inserted into the host color image using the Haar Wavelet Transform, where the copyright of Watermark is printed in [14]. A new robust watermarking scheme is proposed in

      [15], which provides a complete algorithm that embeds and extracts the watermark information effectively.

      In this paper a watermarking technique is proposed, to directly embed multiple watermarks into a single image. The watermark embedding process the multiple watermarks are embedded into original image. The extraction processes recover the watermark from the watermarked image. The experimental results have shown this scheme has preferable performance of imperceptibility. This paper is organized as follows; the suggested technique (Discrete Wavelet Transform) is proposed in section II. The multiple watermarking is explained in section III. The Proposed algorithms for watermark embedding are explained in section IV. The Proposed algorithms for watermark extraction are explained in section V. The discussions are presented in section VI. Finally, concluding remarks are given in section VII.

      (a)

      (b)

      (c)

   2. THE SUGGESTED TECHNIQUE

      An image watermarking algorithm that operates on the wavelet domain is suggested here. It features imperceptibility on the watermarked image as well as robustness of extracted watermark. A gray scale watermark image is inserted into the original color image using the daubechies Wavelet Transform. Fig.1 shows the set of original images 512×512 size of Lena, Boat, Barbara and Peppers as the color images.

      The proposed scheme is based on two-dimensional DWT, each level of produces four bands of data, one corresponding to the approximation sub-band (LL), and three other corresponding to horizontal (HL), vertical (LH), and diagonal (HH) sub-bands. The decomposed image shows an approximation image in the lowest resolution low pass band. The low pass band can further be decomposed to obtain another level of decomposition. Fig. 2 shows the two level decomposition of Lena image.

      Fig. 2. Two level wavelet Decomposition of Lena image

   3. MULTIPLE WATERMARKING

      Most watermarking algorithms support single watermark embedding, but there are great limitations when single watermarking algorithms are tried into practical applications in few rare situation, like when multiple users share the copyright, it is need to support multiple users to embed their watermarks synchronously. This highlights the need for multiple watermarking. To advocate several goals one might wish to embed multiple watermarks into the same image so as to achieve the robustness to large range of image processing operations.

      Fig. 3 shows 32×32 size gray scale logo is used as watermark image, such as watrmark 1 and watermark 2.

      (d)

      Fig. 1. Original Images (a) Lena (b) Boat (c) Barbara (d) Peppers

      (a) (b)

      Fig. 3. (a) Watermark 1 (b) Watermark 2

      784

      Proceedings of International Conference ICSEM13

      The inverse wavelet transform is performed to get the watermarked image. To evaluate the performance of watermarking technique is the peak signal to noise (PSNR). Peak Signal to Noise Ratio (PSNR) is used to measure quality of watermarked image, it is given by

      2552

      PSNR(dB) 10 log10 MSE

      (2)

      Fig. 4. Watermark Embedding Process

      Fig. 5. Watermark Extraction Process

   4. THE EMBEDDING ALGORITHM

      The proposed method embeds multiple watermarks by decomposing the original image. The watermark used for embedding is a binary logo image, which is small compared with the size of the original image. The block diagram of watermark embedding process is shown in Fig. 4.

      The original image is decomposed by 2-levels using DWT. The watermark W1 is embedded in LL2 sub-band and the watermark W2 is embedded in HH2 sub-band. The watermarked image can be obtained by the following equation.

      The TABLE I shows the imperceptibility evaluation of watermarked images on Lena, Boat, Barbara and Peppers images and lists the PSNR value for image quality. Fig. 6. shows comparison of PSNR for various test images.

      TABLE I WATERMARKED IMAGES ON LENA, BOAT, BARBARA AND PEPPERS IMAGES AND ITS IMAGE QUALITY (PSNR)

      Images	Watermarked Images	PSNR in db
      Lena		54.9412
      Boat		54.9391
      Barbara		54.9391

      WI (i, j) I (i, j) w(i, j)

      Where, WI = watermarked image, w = watermark,

      I = cover Image and

      (1)

      = scaling factor which determine the strength of watermark.

      785

      N.Mohananthini,G.Yamuna

      Peppers

      54.3031

      Lena		0.9354		0.9796
      Boat		0.9399		0.9780
      Barbara		0.9348		0.9781
      Peppers		0.9322		0.9787

      PSNR (dB)

      56

      54.9412 54.9391 54.9391

      55 54.3031

      54

      53

      52

      51

      50

      Lena Boat Barbara Peppers

      Fig. 6. Comparison of PSNR for various test images

      1

      0.98

      0.96

      0.94

      0.92

      0.9

      NC Value for

      Watermark 1 Watermark 2

      0.9796 0.978 0.9781 0.9787
      		4			9			8			2
      0.935			0.939			0.934
      									0.932

      Lena Boat Barbara Peppers

   5. THE EXTRACTION ALGORITHM

      The watermark extraction processes are the inverse process of watermark embedding, shown in Fig. 5. Thus the watermark can be recovered exactly from the watermarked image by recovery process.

      The watermarked image and the original image is decomposed by 2-levels, using Discrete wavelet transform. The watermark W1 and W2 can be extracted from the watermarked image sub-bands LL2, and HH2 respectively. Then it is divided by the watermark strength factor . This is summarized as follows,

      Fig. 7. Comparison of NC values for Extracted Watermark 1 and Extracted Watermark 2

   6. DISCUSSIONS

      In this paper, the multiple watermarking scheme is proposed based on discrete wavelet transform for color images. The proposed algorithm uses daubechies wavelet which has the property of regularity when compared to other wavelet transform. The proposed method is two level of decomposition. The multiple watermarks are embedded in original image, to achieve better visual quality on watermarked image. The advantage of the proposed method has preferable performance of

      w' (i, j) (WI (i, j) I (i, j))

      (3)

      imperceptibility and used as the color images of Lena, Boat, Barbara and Peppers images.

      Normalized Correlation is used to measure the quality of watermark after recovery. The NC between the embedded watermark W (i, j) and the extracted watermark W (i, j) is defined as

   7. CONCLUSION

In this paper, we proposed a multiple watermarking scheme on DWT. In the embedding process, the multiple watermarks are embedded to original image. In the extracting process, the original watermark is retrieved

H L from the watermarked image. The proposed method has

NC

W (i, j) W ' (i, j)

i1 j 1

H L

(4)

good imperceptibility on the watermarked image and superior in terms of Peak Signal to Noise Ratio (PSNR). As a future initiative, to achieve a high robustness for

geometric, non geometric and common image processing

i1

j 1

attacks.

The TABLE II shows the extracted watermarks on Lena, Boat, Barbara and Peppers images and lists the values of Normalized Correlation. Fig. 7. shows comparison of NC values for extracted watermark 1 and watermark 2.

TABLE II. EXTRACTED WATERMARKS ON LENA, BOAT, BARBARA AND

PEPPERS IMAGES AND ITS NC VALUES

REFERENCES

1. Q. Ying and W. Ying, A survey of wavelet-domain based digital image watermarking algorithm, Computer Engineering and Applications, 2004, Vol. 11, pp. 46-49.

2. Vaishali S. Jabade and Dr. Sachin R. Gengaje, Literature Review of Wavelet Based Digital Image Watermarking Techniques, International Journal of Computer Applications (0975 8887), Volume 31 No.1, October 2011.

3. R. Safabakhsh, S. Zaboli, A. Tabibiazar, Digital watermarking on still images using Wavelet transform, Proceedings of the International Conference on Information Technology: Coding and

   Computing (ITCC'04), Volume 1, April 05-07, 2004, Las Vegas, Nevada.

4. X. Xia, C. Boncelet, and G. Arce, A multiresolution watermark for digital images, in Proc. IEEE Int. Conf. Image Processing 1997(ICIP'97), Oct. 1997, vol.1, Santa Barbara, pp. 548-551.

5. Ming-Shing Hsieh, Din-Chang Tseng, Hiding Digital Watermarks Using Multiresolution Wavelet Transform, IEEE Transactions on Industrial Electronics, October 2001, VOL. 48, NO. 5.

6. Maha Sharkas, Dahlia R ElShafie, and Nadder Hamdy, A Novel Dual-Purpose Image Watermarking Technique, International Journal of Signal Processing, 2006, Volume 3 Number 4.

7. Peter H. W. Wong, Oscar C. Au, and Y. M. Yeung, A Novel Blind Multiple Watermarking Technique for Images, IEEE Transactions On Circuits And Systems For Video Technology, August 2003, Vol. 13, No. 8.

8. F. Mintzer and G. W. Braudaway, If one watermark is good,are more better?, in Proceedings of the International Conference on Accoustics, Speech, and Signal Processing, Phoenix,Arizona, USA, May. 1999, vol. 4, pp. 20672070.

9. R. Mehul, and R. Priti, Discrete Wavelet Transform Based Multiple Watermarking Scheme, Proceedings of IEEE Region

   10 Technical Conference on Convergent Technologies for the Asia-Pacific, 14 17 October 2003, Bangalore, India.

10. Ibrahim Nasir, Ying Weng, Jianmin Jiang, Novel Multiple Spatial Watermarking Technique in Color Images, IEEE Fifth International Conference on Information Technology: New Generations, 2008.

11. B. Surekha, Dr. GN. Swamy, and Dr. K. Srinivasa Rao, A Multiple Watermarking Technique for Images based on Visual Cryptography, International Journal of Computer Applications (0975 8887), 2010, Volume 1 No. 11.

12. Sarita Kumari and Ritu Vijay, Analysis of Orthogonal and Biorthogonal Wavelet Filters for Image Compression, International Journal of Computer Applications, May 2011, Vol. 21 No.5, pp. 17 19.

13. Piyu Tsai, Yu-Chen Hu, Chin-Chen Chang , A color image watermarking scheme based on color Quantization, Elsevier, Signal Processing, 2004, Vol.84, no.1-2, pp. 95 106.

14. Anuradha, Rudresh Pratap Singh, DWT Based Watermarking Algorithm using Haar Wavelet, International Journal of Electronics and Computer Science Engineering, 2012, Vol. 1, Issue. 1, pp. 01 06.

15. M.Mohamed Sathik and S.S. Sujatha, A Novel DWT Based Invisible Watermarking Technique for Digital Images, International Arab Journal of e-Technology, January 2012, Vol. 2, No.3, pp 167 173.