Robust Image Watermarking Against Different Attacks ... · Robust Image Watermarking Against Different Attacks: Noising ... of papers on document watermarking based on line ... and

Journal of Babylon University/Pure and Applied Sciences/ No.(6)/ Vol.(21): 2013

١٩٧٢

Robust Image Watermarking Against Different Attacks:

Noising, (Jpeg) Compression, And Filtering

Ahmed Hassan Hadi Technical College of Najaf

Abstract Many electronic watermarks for still images are sensitive to geometric distortion. For example, noising, JPEG compression and filtering of an image can prevent blind detection of a public watermark. In this paper, we propose a watermarking algorithm that the watermark is embedded into a (2-D) coefficients obtained by taking DCT of the image is robust to noising, JPEG compression and filtering distortions.

Keywords: Image watermarking, robustness, DCT, watermarking attacks

االخالصة

ضغط والــ علـى سـبیل المثـال الـضوضاء. العدیـد مـن العالمـات المائیـة االلكترونیـة للـصور الثابتـة حـساسة الـى التـشویة الهندسـي

تكــون فیهــا العالمــة فــي هــذا البحــث نقتــرح خوارزمیــة.والمرشــحات للــصورة یمكــن ان تمنــع الكــشف للعالمــة المائیــة) JPEG(باســتخدام

والــضغط باســتخدام متینــة ضــد التــشوهات الحاصــلة نتیجــة الــضوضاءالمائیــة مــضمنة مــع معــامالت تحویلــة الجیــب تمــام الممیــزة وهــي

)JPEG (والمرشحات للصورة.

I. Introduction

The process of embedding a watermark in a multimedia object is termed as watermarking. Watermark can be considered as a kind of a signature that reveals the owner of the multimedia object. Content providers want to embed watermarks in their multimedia objects (digital content) for several reasons like copyright protection, content authentication, tamper detection etc. (Low et al., 1998), (Maxemchuk et al., 1997), (Low et al., 1998), (Low et al., 1995), and (Low et al., 1995) have published a series of papers on document watermarking based on line and word shifting. These methods are applicable to documents that contain paragraphs of printed text. Data is embedded in text documents by shifting lines and words spacing by a small amount (1/150 inch.) For instance, a text line can be moved up to encode a ‘1’ or down to encode a ‘0’, a word can be moved left to encode a ‘1’ or right to encode ‘0’. The techniques are robust to printing, photocopying, and scanning. There are various watermarking applications for images, as listed below (Hartung et al., 1999), and (Seitz et al., 2005). - Copyright protection embeds the copyright owner information in the image. This information is used in order to prevent others from alleging ownership of the image. This is the most common use of watermarking applications currently. - In an image authentication application the intent is to detect alterations to the data. The image’s properties, such as it’s pixel averages, maximums and minimums, are embedded and compared with the current images for modifications. - Prevention of unauthorized copying is accomplished by embedding information about how often an image can be legally copied.

١٩٧٣

- The fingerprint embeds information about the legal receiver in the image. This involves embedding a different watermark into each distributed image and allows the owner to locate and monitor pirated images that are illegally obtained. (Hwang, et al., 2000) proposed most significant bit watermarking embedding (MWE) scheme that uses visual cryptography for generation of shares based on the pixel value of the binary watermark and most significant bit of pixel value of the image (Hwang, et al., 2000), (Sleit, et al., 2008), and (Surekha, et al., 2011). There are few algorithms that modify these DFT magnitude and phase coefficients to embed waternarks. (Ruanaidh et al., 1996) proposed a DFT watermarking technique in which watermark is embedded by modifying the phase information within the DFT. It has been shown that phase based watermarking is robust against image contrast operation. Later (Ruanaidh et al., 1998) showed how Fourier Mellin transform can be used for digital watermarking. Fourier Mellin transform is similar to apply Fourier transform to log-polar coordinate system for an image. This scheme is robust against RST attacks. (De Rosa et al., 1999) propose a scheme to insert watermark by directly modifying the mid frequency bands of the DFT magnitude component. (Ramkumar et al., 1999) also present a data hiding scheme based on DFT, where they modify the magnitude component of the DFT coefficients. Their simulations suggest that magnitude DFT survives practical compression, and this can be attributed to the fact that most practical compression schemes try to maximize the PSNR. Hence using magnitude DFT is a way to exploit the hole in most practical compression schemes. The proposed technique is shown to be resistant to JPEG and SPIHT compression. (Lin et al., 2001) present a RST resilient watermarking algorithm. The watermark is embedded in the magnitude coefficients of the Fourier transform re-sampled by log-polar mapping. The technique is however not robust against cropping and shows weak robustness against JPEG compression (QF 70). There is no a universal set of requirements that all watermarking technique must satisfy. Nevertheless, some general directions can be given for most of the applications (Podilchuk et al., 2001), (Cox et al., 2002), and (Katzenbeisser et al., 2000): (1) Perceptual transparency: in most applications, embedding a watermark should not affect the quality of the original media. (2) Robustness: watermarking can be robust, fragile and semi-fragile. Robust watermarking is employed to protect and verify the copyright; the purpose of fragile watermarking is content authentication for integrity and reality; and semi-fragile is used to distinguish between “information preserving” lossy transformations, such as compression, and “information altering” lossy transformations, such as replacement of objects (Podilchuk et al., 2001). Robustness is an important parameter for the watermarking detector with different strength. For example, a watermarking scheme is defined as robust if the detector can successfully detect the watermark in the processed image (Piper et al., 2009). Also, more precisely, robustness can be defined as the degree of resistance of a watermarking scheme to modifications of the host signal due to either common signal processing, or operations devised specifically in order to render the watermark undetectable (Tefas et al., 2009). (3) Capacity: the amount of watermark bits that can be stored in the host media. (4) Security: two levels of security can be identified (Katzenbeisser et al., 2000). At the higher level of security, an unauthorized party can neither detect the existence of


١٩٧٤

the embedded watermark, nor extract it. At the lower level, people can detect if the media are watermarked, however, the watermark cannot be extracted.

II. Insertion and Detection Algorithm Figure (1) shows the insertion and detection watermarking algorithm. Insertion Algorithm

1. Reading an image with size (512×512). 2. Take (256 × 256) discrete cosine transform of image size (512×512). 3. Take any two letters as the watermark 4. Convert the letters to gray level image with size (256×256). 5. multiply the watermark image with factor and the result plus one (x = w + 1), let = 10. 6. multiply the new watermark by dct's coefficients [by using block process

in MATLAB] 7. Take (256 × 256) inverse discrete cosine transform for coefficient in step

6. 8. Image watermarked applied to one of attacks "noise or compression

(JPEG) or filtering"

Detection Algorithm

1. Take (256 × 256) discrete cosine transform of watermarked image. 2. Take (256 × 256) discrete cosine transform of original image. 3. divide coefficients of step1 on coefficients of step2 to get watermark as image

(512×512).

١٩٧٥

IDCT

Figure (1) The proposed watermarking algorithm

Convert letters to gray level

image "W"

Any

watermark Letters as TV

New Watermark

" W+1"

Embedding

Original Image 512512

Block Process DCT2

Watermarked Image

After attacks

Extracted Watermark Image

Attacks (Noise or compression or filtering)

Watermarked Image


١٩٧٦

III. Experimental results A gray image of size 512×512 is analyzed using discrete cosine transform (DCT), the watermark is embedded in the DCT coefficients. and results are given in figures below. The watermark that is embedding in images is shown in Figure (2). the experiments using the gray-level images and watermarked images shown in Figure (3). Figure (4), and (7) shows the effect of an attack based on worst case noising and its extracted watermark for each case. The results for JPEG compression are displayed in Figure (5), and (8) for different quality factor (50%, 20%, 10%) and the watermark is detected even for a quality factor10%. Figure (6), and (9) show the results for unsharp, negative values, and wiener filtering, respectively. That applied on the watermarked images to attempt to destroy the watermark , and the proposed algorithm can extract the watermark for each case. Robust against adding noise, robust against JPEG compression, and robust against filters illustrated as shown in Table (1) Table (1) Robust results against adding noise, compression and filtering

Attacks

Parameter

Adding noise

0.2%, 2%, and 10%

Compression JPEG

50%, 20%, and 10%

Filtering

Unsharp, negative values, Wiener

١٩٧٧

Figure (2) Watermark

(a) (b)

(c) (d) Figure (3) An original image a, c (left column) and its watermarked image b, d (right column).


١٩٧٨

Exp.1: Robust against adding noise

(a) Adding noise, 0.2% and its extracted watermark

(b) Adding noise, 2% and its extracted watermark

(c) Adding noise, 10% and its extracted watermark

Figure (4) Adding noise, (a) 0.2%, (b) 2%, (c)10% and its extracted watermark for each case.

١٩٧٩

Robust against JPEG compression

(a) JPEG compression 50% and its extracted watermark

(b) JPEG compression 20% and its extracted watermark

(c) JPEG compression 10% and its extracted watermark

Figure (5) JPEG compression, (a)50%, (b)20%, (c)10%, and its extracted watermark for each case.


١٩٨٠

Robust against filters

(a) Unsharp filter and its extracted watermark.

(b) Negative values filter and its extracted watermark.

(c) Wiener filter and its extracted watermark.

Figure (6) Filters, (a) Unsharp, (b) Negative values, (c) Wiener, and its extracted watermark for each case.

١٩٨١

Exp.2: Robust against adding noise (a) Adding noise, 0.2% and its extracted watermark 0.02 (b) Adding noise, 2% and its extracted watermark (c) Adding noise, 10% and its extracted watermark Figure (7) Adding noise, (a) 0.2%, (b) 2%, (c) 10% and its extracted watermark for each case.


١٩٨٢

Robust against JPEG compression

(a) JPEG compression 50% and its extracted watermark

(b) JPEG compression 20% and its extracted watermark

(c) JPEG compression 10% and its extracted watermark Figure (8) JPEG compression, (a)50%, (b)20%, (c)10%, and its extracted watermark for each case.

١٩٨٣

Robust against filters

(a) Unsharp filter and its extracted watermark.

(b) Negative values filter and its extracted watermark.

(c) Wiener filter and its extracted watermark.

Figure (9) Filters, (a) Unsharp, (b) negative values, (c) Wiener, and its extracted watermark for each case.


١٩٨٤

Conclusion This paper describes an algorithm for recovering watermark as image ("512 × 512" pixel). Any two letters convert to image ("256 × 256" pixel) as watermark is embedded in the DCT coefficients to get invisible watermarking for image protection. Experimental results demonstrate that this algorithm is robust against spatial attacks including noising, JPEG compression and filtering.

References

Cox I.J., Miller M.L., Bloom J.A., (2002), " Digital Watermarking," Morgan Kaufmann, San Francisco.

De Rosa A., Bami M., Bartolini F., Cappellini V., Piva A., (1999), "Optimum Decoding of Non-additive Full Frame DFT Watermarks," In procedings of the 3rd Workshop of Information Hiding, pp. 159-171.

Hartung F., and Kutter M., (1999), "Multimedia watermarking techniques," Proceedings of the IEEE, Vol. 87, No. 7, pp. 1079–1107.

Hwang R-J., (2000), "A digital image copyright protection scheme based on visual cryptography," Tamkang journal of Science and Engineering, Vol. 3, No. 2, pp. 97-106.

Katzenbeisser S., Petitcolas F.A.P. (Eds.), (2000), " Information Hiding Techniques for Steganography and Digital Watermarking," Artech House, Boston, London.

Low S. H., Lapone A. M., and Maxmchuk N. F., (1995), "Document identification to discourage illicit copying," IEEE GlobeCom 95, Nov. 13-17, Singapore.

Low S. H., and Maxemchuk N. F., (1998), "Performance comparison of two text marking methods," IEEE Journal on Selected Areas in Communications, Vol. 16, No. 4, pp. 561-572.

Low S. H., Maxemchuk N. F., Brassil J. T., and O’Gorman L., (1995), "Document marking and identification using both line and word shifting," Infocom 95, IEEE Computer Society Press, Los Alamitos, Calif., Vol. 2, pp. 853-860.

Low S. H., Maxemchuk N. F., and Lapone A. M., (1998), "Document identification for copyright protection using centroid detection," IEEE Trans. on Comm., Vol. 46, No. 3, pp. 372-83.

Lin C-Y, Wu M., Bloom JA., Cox U., Miller ML., and Lui YM., (2001), "Rotation, Scale and Translation Resilient Watermarking for Images," IEEE Transactions on Image Processing, Vol. 10, No. 5, pp.767-782.

Maxemchuk N. F., and Low S. H., (1997), "Marking text documents," Proc. IEEE Intl Conf. On Image Processing.

Piper A. and Safavi-Naini R., (2009), "How to Compare Image Watermarking Algorithms," in Transactions on Data Hiding and Multimedia Security IV. Vol. 5510, Y. Shi, Ed., ed: Springer Berlin / Heidelberg, pp. 1-28.

Podilchuk C.I., Delp E.J., (2001), "Digital watermarking: algorithms and applications," IEEE Signal Process. pp. 33–46.

Ramkumar M., Akansu A.N., Alatan A.A., (1999), "A Robust Data Hiding Scheme for Digital Images Using DFT," in IEEE ICIP, Vol. 2, pp. 211-215.

Ruanaidh J. J. K. O., Dowling W. J., Borland F. M., (1996), "Phase watermarking of digital images," in Proc. IEEE Int. Conf: Image Processing, pp. 239-242, Sept. 16-19.

Ruanaidh J. J. K. O., Pun T., (1998), "Rotation, scale and translation invariant spread spectrum digital image watermarking," Signal Process, Vol. 66, No. 3, pp. 303-317.

١٩٨٥

Seitz J., (2005), "Digital Watermarking for Digital Media". Information Science Publishing.

Sleit A., and Abusitta A., (2008), "A visual cryptography based watermark technology for individual and group images," Systems, Cybernetics and Informatics, Vol. 5, No. 2, pp.24-32.

Surekha B., and Swamy G.N., (2011), "A spatial domain public image watermarking," International Journal of Security and Applications, Vol. 5, No. 1, pp. 1-11.

Tefas A., Nikolaidis N., and Pitas I., (2009), "Chapter 22 – Image Watermarking: Techniques and Applications," in The Essential Guide to Image Processing (Second Edition), B. Al, Ed., ed Boston: Academic Press, pp. 597-648.

Documents

Robust Image Watermarking Against Different Attacks ... · Robust Image Watermarking Against Different Attacks: Noising ... of papers on document watermarking based on line ... and