EE 5359 FALL 2010 DIGITAL WATERMARKING. Abrar Ahmed Syed Under the guidance of Dr. K. R. Rao

Transcription

1 EE 5359 FALL 2010 DIGITAL WATERMARKING Abrar Ahmed Syed Under the guidance of Dr. K. R. Rao

2 KEY POINTS What is digital watermarking? Why do we need digital watermarking? What are the different types of watermarking? What are its characteristics and requirements? What are the different techniques and schemes used to watermark? What are the different types of attacks it is susceptible to? What are the ways of counter-attacking a watermarking attack?

3 PROJECT GOAL The goal of this project can be broadly classified into six steps: 1. Implementing two forms of watermarking techniques- Visible & Invisible 2. Implementing an attack on the visible watermark image (For the sake of this project, adding noise is used as a means of attack) 3. The watermarked image with noise added to it is then compressed and decompressed using JPEG compression technique. This acts as another form of attack. 4. The next step is to remove the noise from the decompressed watermarked image. 5. Next, seperating both the base and the watermark images from the concatenated watermarked image. 6. Finally, a benchmarking of original and recovered images is done based on PSNR, SSIM and SSIM map.

4 DIGITAL WATERMARKING As advances are made in field of communication, it became necessary to cipher and decipher. This led to discovery of stenography and watermarking. Stenography: Hiding information over cover. Watermarking: Hiding information related to cover. Embedding is done by manipulating contents in the signal itself and is made imperceptible.

5 DIGITAL WATERMARKING Embedding a digital signal (audio, video or image) with information which cannot be removed easily is called digital watermarking. Figure 1 shows the block diagram of embedding digital watermark. Signal Attacking function Signal Any embedding function E Detecting and retrieving function Figure 1: Block diagram of embedding digital watermark

6 EMBEDDING For example sake, figure 2 shows the block diagram for audio watermark encoding. Original signal Framing Spectral analysis Watermarked signal Watermark addition DC Carrier removal Figure 2: Encoding block diagram of audio watermarking technique

7 RETREIVING For example sake, figure 3 shows the block diagram of audio watermark decoding. Watermarked signal Framing Spectral analysis Original signal Watermark processing Figure 3: Decoding block diagram of audio watermarking.

8 APPLICATIONS Ownership assertion: Watermarking is used to establish ownership of content. Fingerprinting: Watermarking is used to avoid illegal distribution of media. Authentication and integrity verification: Content which is protected by key verification should not be accessible without authentication. Content labeling: Bits embedded in data giving extra information Usage control: To limit copies creation of copyrighted data, by blocking using watermark. Content protection: Visible watermark block is used for this purpose. No universal technique to satisfy all of these. (Source: [23] N. Memo and P. W. Wong, Protecting digital media content, Communications of the ACM, Vol. 41, pp , 1998)

9 CLASSIFICATIONS Visible Invisible Robust Fragile Public and Private Capacity Perceptibility Embedding techniques Spread spectrum Quantization Amplitude modification

10 TECHNIQUES OR SCHEMES OF WATERMARKING Spatial domain techniques Least significant bit coding (LSB) Predictive coding schemes Correlation-based techniques Patchwork techniques Frequency domain techniques Discrete cosine transform (DCT) based technique Wavelet transform based watermarking Simple watermarking

11 ATTACKS Basic Robustness Presentation Interpretation Implementation Removal Geometrical Cryptographic Active & Passive Forgery Collusion Distortive

12 COUNTER ATTACKS Power spectrum condition Noise visiblity function

13 PROJECT PART 1(INVISIBLE) From slide 3, the part 1 of this project is: Implementing two forms of watermarking techniques- Visible & Invisible Invisible Watermarking An image A will be selected as base image. An image B will be selected as watermark image. The MSB of B will be read and these will be written on the LSB of A. Thus, A will be watermarked with B resulting in a combined image C. C therefore will now contain an image A which has its LSB s replaced with the MSBs of B. The technique used will be LSB technique which is a form of spatial domain technique.

14 PART 1 (INVISIBLE) Image A and watermark B are read. MSBs of B will be are transferred to the LSBs of A Watermarked image C, contains MSBs of B and LSBs of A Figure 4: Block diagram of invisible watermarking.

15 PART 1 (INVISIBLE) PROGRAM 1. base_image = input('enter base image file name with extension : ', 's'); 2. watermark_image = input('enter watermark image file name with extension: ', 's'); 3. base = imread(base_image); 4. watermark = imread(watermark_image); 5. figure(1) 6. imshow(base,[]) 7. title('original Image (Cover Image/Base Image)') 8. figure(2) 9. imshow(watermark,[]) 10. title('watermark Image (Image which will hide in cover image)') 11. base=double(base); 12. watermark=double(watermark); 13. bits=1; 14. watermark_shifted=bitshift(watermark,-(8-bits)); 15. for i=1:bits 16. base=bitset(base,i,0); 17. End 18. watermarked_image = uint8(base+watermark_shifted); 19. figure(3) 20. imshow(watermarked_image,[]) 21. title('watermarked Image (Final watermarked image)') 22. imwrite(watermarked_image,'invsible watermarked image.bmp'); % Save modified image

16 PART 1 (INVISIBLE) EXPLANATION s

17 PART 1 (INVISIBLE) EXPLANATION

18 PART 1 (INVISIBLE) EXPLANATION

19 PART 1 (INVISIBLE) EXPLANATION

20 PART 1 (INVISIBLE) RESULT Figure 5: Base Image Figure 6: Watermark Image Figure 7: Watermarked Image

21 PROJECT PART 1(VISIBLE) From slide 3, the part 1 of this project is: Implementing two forms of watermarking techniques- Visible & Invisible Visible Watermarking A raw bitmap base image A will be selected from the set of standard test images. A raw bitmap watermark image B will be selected from the set of standard test images. Now both images A and B will be concatenated to get a watermarked image D D therefore will now contain the base image A and the watermark image B. This technique is used to add visible watermark in the image.

22 PART 1 (VISIBLE) Image A and watermark B are read and displayed Zero matrix initialized with dimensions 512x512 The first column of pixels of the zero matrix is filled with the first column of pixels of base image Watermarked image thus obtained is displayed Process repeated for all 256 column of base and watermark images respectively Figure 8: Block diagram of visible watermarking technique The second column of pixels of the zero matrix is filled with the second column of pixels of watermark image

23 PART 1 (VISIBLE) PROGRAM 1. base_image = input('enter base image file name with extension : ', 's'); 2. watermark_image = input('enter watermark image file name with extension: ', 's'); 3. base = imread(base_image); 4. watermark = imread(watermark_image); 5. figure(1) 6. imshow(base,[]) 7. title('original Image (Cover Image/Base Image)') 8. figure(2) 9. imshow(watermark,[]) 10. title('watermark Image (Image which will hide in cover image)') 11. base=double(base); 12. watermark=double(watermark); 13. realdata = zeros([ ]);

24 PART 1 (VISIBLE) PROGRAM 14. indexi=1; 15. indexj=1; 16. realindexi = 1; 17. realindexj = 1; 18. for i=1: for j=1: realdata(realindexi,realindexj)=base(indexi,indexj); 21. indexj=indexj+1; 22. realindexj=realindexj+2; 23. end 24. indexi= indexi+1; 25. indexj = 1; 26. realindexi=realindexi+2; 27. realindexj=1; 28. End

25 PART 1 (VISIBLE) PROGRAM 29. indexi=1; 30. indexj=1; 31. realindexi = 2; 32. realindexj = 2; 33. for i=1: for j=1: realdata(realindexi,realindexj)=watermark(indexi,indexj); 36. indexj=indexj+1; 37. realindexj=realindexj+2; 38. end 39. indexi= indexi+1; 40. indexj = 1; 41. realindexi=realindexi+2; 42. realindexj=1; 43. End 44. realdata =uint8(realdata); 45. figure(3) 46. imshow(realdata,[]) 47. title('watermarked Image (Final watermarked image)') 48. imwrite(realdata,'visible Watermarked.bmp'); % Save modified image

26 PART 1 (VISIBLE) PROGRAM To better understand the concatenation process, depicting the original base image as a block Fig.9 and original watermark image as a block Fig.10. The original blocks, or images in our case are 256X256. The base image will have X elements and the watermark image have Y elements. + Figure 9: Base image Figure 10: Watermark Figure 11: Zero Matrix Figure 12: Watermarked Now a zero matrix, Fig.11, of 512x512 size is initialized. This is the double the size of our base and watermark images which are 256x256 in size. Now in the program, the base image is concatenated with the watermark image to form the watermarked image. This is done by reading the first element of the base matrix and writing it in the first element of the zero matrix. It then reads the second element of base matrix and instead of writing it in the second element of the zero matrix, it writes the base element into the third element of the zero matrix, thus skipping one space in between. This step is repeated till all the elements of the base matrix are written in the zero matrix (i.e. by skipping the alternate rows and alternate columns). Similarly the watermark is added in the next row in alternate columns. The remaining elements are kept zero. That is the reason why the watermark has a black colored mesh like structure on it. Thus a concatenated matrix of both base and watermark images is obtained at the end of execution of the whole program

27 PART 1 (VISIBLE) RESULT Figure 13: Base image Figure 14: Watermark Image Figure 15: Watermarked Image

28 PROJECT PARTS 2&3 (ATTACKS) From slide 3, the parts 2&3 of this project are: Implementing an attack on the visible watermark image (For the sake of this project, adding noise is used as a means of attack) The watermarked image with noise added to it is then compressed and decompressed using JPEG compression technique. This acts as another form of attack. Noise Addition Noise is added to concatenate watermarked image D. The noise is a random matrix of order [512x512]. The resulting image after the addition of noise is displayed and stored as Noisy Image, E.

29 PROJECT PARTS 3&4(ATTACKS) JPEG Compression & Decompression Once the noise is added, the program halts and waits for the JPEG compression to be done Once the compression is done, the program is informed by pressing enter key that the compression is done and it can proceed by removing the noise. JPEG Compression is done by running the.exe file available at [19] The watermarked image added with noise which is in bitmap format is compressed into JPEG format. The size of the original image is 257KB. The JPEG compressed image is 92.2KB. The 4 th step is denoising The JPEG compressed image, is now denoised. The program which was on hold is triggered by pressing the Enter Key. The program runs by reading the JPEG compressed noised and watermarked image. The noise is now removed from the JPEG compressed image.

30 PROJECT PARTS 2&3 (ATTACKS) Random matrix of 512x512 dimensions defined as Noise matrix Noise added to the watermarked image Figure 16: Block diagram showing noise addition Program awaits users input. If Jpeg compression done, press Enter When triggered, the program continues executing the rest of the implementation Figure 17: Block diagram showing JPEG compression

31 PROJECT PARTS 3&4 (ATTACKS) JPEG COMPRESSION Done by running cjpeg.exe available at [19] Figure 18: Block diagram showing JPEG decompression Bitmap image is compressed to JPEG image The output by default is.gif, it is made to.jpg using outfile command Image file obtained is of smaller size, since it is JPEG JPEG image is decompressed using djpeg.exe After decompression, standard noise matrix initialized earlier is removed The watermarked image is recovered Figure 19: Block diagram showing noise removal

32 PARTS 2,3 & 4 (ATTACKS) PROGRAM imdata = imread('visible Watermarked.bmp'); noisy = rand([ ]); imdata1= double(imdata) + double(noisy); imdata1 = uint8(imdata1); figure(4) imshow(imdata1,[]) title('watermarked image with noise added to it') imwrite(imdata1,'noisy.bmp ) question = input('if you have completed the JPEG compression of image press enter: ', 's'); imdata2=imread('noise.jpg'); imdata2 = imdata2(:,:,1); imdata3 = double(imdata2) - double(noisy); imdata3 = uint8(imdata3); figure(5) imshow(imdata3,[]); title('recovered image from noise') imwrite(imdata3,'watermarked Image Recovered.bmp')

33 PARTS 2,3&4 (ATTACKS) RESULT Figure 20: Watermarked image Figure 21: Watermarked image with noise added Figure 22: Recovered image from noise

34 PROJECT PART 5(RECOVERING) From slide 3, the part 5 of this project are: 1. The next step is to separate both the base and the watermark images from the concatenated watermarked image. Seperating the images 1. Finally the base image A and the watermark image B are separated from the concatenated recovered watermarked image G 2. The images are in JPEG format, when compared to the original images, which were in Bitmap formats. Therefore their sizes are less.

35 PROJECT PART 5(RECOVERING) Concatenated watermarked image is recovered The base image and watermark image are finally seperated The images obtained are in JPEG format, since the concatenated image was JPEG compressed and decompressed Figure 23: Block diagram of recovering images from watermark Base image & watermark image are displayed

36 PART 5 (RECOVERING) PROGRAM 1. imdata3=double(imdata3); 2. realindexi = 1; 3. realindexj = 1; 4. for i=1: for j=1: base(i,j)=realdata(realindexi,realindexj); 7. realindexj=realindexj+2; 8. end 9. realindexj=1; 10. realindexi=realindexi+2; 11. End 12. realindexi = 2; 13. realindexj = 2; 14. for i=1: for j=1: watermark(i,j)=realdata(realindexi,realindexj); 17. realindexj=realindexj+2; 18. end 19. realindexj=1; 20. realindexi=realindexi+2; 21. end

37 PART 5 (RECOVERING) PROGRAM 22. base=uint8(base); 23. watermark=uint8(watermark); 24. figure(6) 25. imshow(base,[]) 26. title('recovered Base Image at the End') 27. imwrite(base,'rbi.jpg'); % Save modified image 28. igure(7) 29. imshow(watermark,[]) 30. title('recovered Watermark Image at the End') 31. imwrite(watermark,'rwi.jpg );

38 PART 5 (RECOVERING) RESULT Figure 24: Recovered watermark image from noise Figure 25: Recovered base image Figure 26: Recovered watermark image

39 PROJECT PART 6 (BENCHMARKING) From slide 3, the part 6 of this project are: 1. Finally, a benchmarking of original and recovered images is done based on PSNR, SSIM and SSIM map. Base image (Comparision between recovered and original base images) Original base image size=65kb Original base image format=bitmap Recovered base image size=19kb Recovered base image format=jpeg PSNR= SSIM= SSIM MAP Figure 27: SSIM map of base image

40 PROJECT PART 6 (BENCHMARKING) Watermark image (Comparision between recovered and original watermark images) Original watermark image size=65kb Original watermark image format=bitmap Recovered watermark image size=20.2kb Recovered watermark image format=jpeg PSNR= SSIM= SSIM MAP Figure 28: SSIM map of watermark image

41 CONCLUSIONS Using Matlab, two forms of watermarking techniques are implemented. One being invisible and the other being visible. Noise is added to the images as a form of attack. The images are compressed and decompressed as another form of attack. The noise is later removed and the base and watermark images are separated from the watermarked image. Finally, a benchmarking of original and recovered images is done based on PSNR, SSIM and SSIM map.

42 REFERENCES [1] G. K. Wallace, The JPEG still picture compression standard, IEEE Trans. on Consumer Electronics, Vol. 38, pp.18-34, Feb [2] R. Popa, An analysis of steganographic techniques, The Politehnica University of Timisoara, Faculty of Automatics and Computers, Department of Computer Science and Software Engineering, Website: [3] P. Vidyasagar, S. Han and E. Chang. "A survey of digital image watermarking techniques, 3rd IEEE International Conference on Industrial Informatics (INDIN 2005), edited by T. Dillon, X. Yu. and E. Chang, pp , Perth, Western Australia, [4] J. Dugelay and S. Roche, A Survey of Current Watermarking Techniques in Information Techniques for Steganography and Digital Watermarking, S.C. Katzenbeisser et al, Eds. Northwood, MA: Artec House, pp , Dec [5] I. J. Cox, et al, "Digital watermarking and steganography" (Second Edition), Morgan Kaufmann, [6] K. R. Rao and P. Yip, Discrete Cosine Transform: Properties, Algorithms, Advantages, Applications, Academic Press, Massachusetts, [7] A. Khan and A.M. Mirza, Genetic perceptual shaping: utilizing cover image and conceivable attack information during watermark embedding. Inf. Fusion, Vol. 8, pp , Oct [8] T. C. Lin and C. M. Lin, Wavelet based copyright protection scheme for digital images based on local features, Information Sciences: an International Journal, Vol. 179, Sept

43 REFERENCES [9] N. Ahmed, T. Natarajan, and K. R. Rao, "Discrete cosine transform", IEEE Trans. Computers, Vol. 23, pp , Jan [10] Prime Core Solutions, website: [11] I.J. Cox, M.L. Miller, J.M.G. Linnartz and T. Kalker, A Review of Watermarking Principles and Practices in Digital Signal Processing for Multimedia Systems, K.K. Parhi and T. Nishitani, New York, Marcel Dekker, New York, pp , [12] S. F. Marseken L. M. Surhone and M. T. Timpledon, Short-time Fourier transform: Window Function, Complex Analysis, Modified Discrete Cosine Transform, Wavelet Series, Betascript Publishers, [13] G. Langelaar, I. Setyawan and R.L. Lagendijk, Watermarking Digital Image and Video Data, in IEEE Signal Processing Magazine, Vol. 17, pp , Sept [14] H. Inoue, A. Miyazaki and T. Katsura, An Image Watermarking Method Based on the Wavelet Transform, IEEE Conf. on Image Processing, Vol. 1, pp , [15] J. Cummins, P. Diskin, S. Lau and R. Parlett, Stegnography and digital watermarking, School of Computer Science, The University of Birmingham, Website: aphy.pdf [16] M. Yeung, B. Yeo and M. Holliman, Digital watermarks: shedding light on the invisible, IEEE Micro, Vol. 18, pp Nov [17] K. S. Jonathan and B. Girod, Power-Spectrum Condition for Energy-Efficient Watermarking, IEEE Trans. on Multimedia, Vol. 4, pp. 551, Dec [18] Privacy principles for digital watermarking, Ver. 1.0, May Website:

44 REFERENCES [19] [20] X. Shang, Structural similarity based image quality assessment: pooling strategies and applications to image compression and digit recognition, M.S. Thesis, EE Department, The University of Texas at Arlington, Aug [21] Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, "Image quality assessment: From error visibility to structural similarity", IEEE Trans. on Image Processing,Vol. 13, no. 4, pp , Apr [22] Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, "Image quality assessment Matlab Program Website: [23] N. Memon and P. W. Wong, Protecting digital media content, Communications of the ACM, Vol. 41, pp , Feb [24] J. S. Pan, H. C. Huang and L. C. Jain, Intelligent watermarking techniques - series on innovative intelligence, Vol. 7, World Scientific Publication, [25] S. Voloshynovskiy, S. Pereira and T. Pun, Attacks on digital watermarks: classification, estimation-based attacks and benchmarks, IEEE Comm. Mag., Vol. 39, pp. 2 10, Aug [26] X. Jian-hui, W. Li-na and Z. Huan-guo, Wavelet based denoising attack on image watermarking, Wuhan University Journal of Natural Sciences, Vol.10, pp , Oct [27] M. Arnold, M. Schmucker and S. D. Wolthusen. Techniques & Application of Digital Watermarking and Content Protection, Artech House Publications, Boston, London, 2003

45 REFERENCES [28] F. Jin, P. Fieguth, L. Winger, and E. Jernigan, Adaptive Wiener image filtering of noisy images and image sequences, Proc. IEEE Int. Conf. on Image Process, pp , Sept [29] F. Mintzer and G. W. Braudaway, If one watermark is good or more better?, IEEE, Int. Conf. on Acoustics, Speech and Signal Processing- ICASSP, pp , [30] Electrical and Computer Engineering, University of Victoria, British Columbia. Website: [31] R.C. Gonzalez and R.E. Woods, Digital Image Processing, Prentice Hall Inc., Upper Saddle River, New Jersey, [32] J.R. Hernandez, M.Amado, and F. Perez-Gonzalez, DCT-Domain Watermarking Techniques for Still Images: Detector Performance Analysis and a New Structure, in IEEE Trans. Image Processing, Vol. 9, pp , Jan [33] S.C. Katzenbeisser, Principles of Steganography in Information Techniques for Steganography and Digital Watermarking, S.C. Katzenbeisser et al, Eds. Northwood, MA, pp. 2-40, Dec [34] M. Kutter and F. Hartung, Introduction to Watermarking Techniques in Information Techniques for Steganography and Digital Watermarking, S.C. Katzenbeisser et al, Eds. Norwood, MA, pp , Dec [35] Bytescout watermarking software. Website: