Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
i Jm Pict 040203

i Jm Pict 040203

Ratings: (0)|Views: 1 |Likes:
Published by ijmpict
A DWT, DCT AND SVD BASED WATERMARKING
TECHNIQUE TO PROTECT THE IMAGE PIRACY
A DWT, DCT AND SVD BASED WATERMARKING
TECHNIQUE TO PROTECT THE IMAGE PIRACY

More info:

Categories:Types, Research
Published by: ijmpict on Jul 12, 2013
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

04/03/2014

pdf

text

original

 
International Journal of Managing Public Sector Information and Communication Technologies (IJMPICT)Vol. 4, No. 2, June 2013
DOI : 10.5121/ijmpict.2013.420321
 ADWT,DCT
 AND
SVDB
 ASED
 W
 ATERMARKING
 T
ECHNIQUE TO
P
ROTECTTHE
I
MAGE
P
IRACY 
Md. Maklachur Rahman
1
1
Department ofComputer Science and Engineering, Chittagong University of Engineering and Technology, Bangladesh
 A
 BSTRACT 
With therapid development of information technology and multimedia, the use of digital data is increasingday by day. So itbecomesvery essential to protectmultimedia informationfrom piracy and also it ischallenging.A great deal of Copyright owners isworriedaboutprotecting anykind ofillegalrepetitionof theirinformation.Hence,facingallthesekinds of problemsdevelopment of the techniques is veryimportant. Digital watermarkingconsideredas a solutionto prevent themultimedia data. In this paper,anideaofwatermarkingis proposed and implemented. Inproposed watermarking method,the original image is rearranged using zigzag sequence andDWT is applied on rearranged image. Then DCT and SVD are applied on all high bandsLH, HL and HH.Watermarkisthen embeddedby modifyingthesingular values of these bands. Extraction of watermark isperformed bytheinversionof watermark embedding process.For choosingof these three bands itgives facility ofmid-band andpure high band that ensuresgoodimperceptibility andmore robustness against different kinds ofattacks.
 K 
 EYWORDS
 DWT, DCT, SVD,Watermarking, Zigzag.
1.I
NTRODUCTION
In recent years, the increasing amountof applicationsusing digital multimediatechnologies hasemphasizedthe need to protect digital multimedia data from pirates.Authentication andinformation hiding, copyrightprotection, contentidentificationandproof ownership havealsobecome important issues. Toaccomplishthese issues, watermarkingtechnology isused.Researchersareinterestedin the field ofwatermarkingbecause ofitssignificance.These kinds of work in this field havelead to several watermarking techniques such asspatial domain andtransformdomain. In transform domain itmaydiscretecosine transform (DCT), discrete wavelettransform (DWT), singular value decomposition (SVD)andtheir crossrelation.Watermarkingisa processembedding a piece of information into amultimedia content, such as image,audio andvideoin such a waythatit is imperceptible to a human, but easily detectable by computer. Beforethedevelopmentofdigital image watermarking,itwasdifficult to achieve copyrightprotection,authentication,datahiding, content identificationandproof ownership. Butcurrentlyit is easy toaccomplishthesekinds goalusing watermarking techniques. So watermarking is very importantto us for these kinds of work.Every watermarking algorithmconsists ofan embedding andextraction process that needstohide desired information.The embedding and extraction processare described below. Embedded watermarkmayhave several properties such as imperceptibilityand robustness. If we cannot distinguish between host image and watermarked imagecalledimperceptibility. Basically imperceptibility depends on similarity between host image andwatermarked image. If it difficult to remove or destroy watermark from watermarked image thenit said to be robustness. Robustness measures how difficult to remove or destroy watermark fromwatermarked image. If it is high then robustness is high.
DCTbased watermarkingcontainsthe low frequencyinformation soimage containsall informationthatissimilar to the
 
International Journal of Managing Public Sector Information and Communication Technologies (IJMPICT)Vol. 4, No. 2, June 2013
22
originalimage.DWTbased compression offers scalability so image can bedividedinto
four sub bands in every level of decomposed and by choosing ofthe sub-bandtodevelop a hybridwatermarkingscheme forimprovingtherobustness, imperceptibility and capacityand help todevelop a new hybrid method.In the first method (ref. 9) the original image is segmented intoblocks then find outspatial frequency of each block select reference image undercertaincondition then apply DWT, DCT and SVD transformation.In the second method (ref. 3) theoriginalimage is segmented into blocks then find out spatial frequency of each block selectreference image under certain condition then apply DWTand SVD transformation.In theproposed method,sincehigh band is considered so it fulfills the requirementsimperceptibilityand robustness.Most of the domaintransformation watermarking techniqueworkswith DCT,DWT, SVD and their mixing algorithm such as DWT-DCT, DCT-SVD and so on.Inthis paperwe proposed adigital watermarkingtechniqueusingDWT-DCT andSVDtransformation. Thismethod isprovided a good imperceptibility and high robustness againstvarious kindsprocessingattacks.Therest ofthe paper is organized as follows:Section2, focusesonoverview of transforms for watermarking. Section 3,gives details theproposed methodologyandwatermarkingalgorithms. In section 4,gives experimental results andcompares. Insection 5,conclusion is drawn.
2. P
RELIMINARIES
Asstatedearlier that transform domain based watermarkingscheme is always abetter choice thanspatial domainbasedwatermarkingscheme. Thiscan be doneby using different transformationlikeDCT, SVD andDWT.In this section, we will briefly describetheDCT, DWT and SVDtransformationsin below.
2.1. Discrete Wavelet Transform (DWT)
The basic ideaofDWTin whicha one dimensional signalisdividedintwo partsone ishighfrequency partandanotheris low frequencypart.Then the low frequency part is split into twoparts andthe similarprocess will continue until the desiredlevel.The high frequency partof thesignal is contained by the edge componentsof the signal.In each level of theDWT (DiscreteWavelet Transform)decomposition an imageseparatesinto fourparts these areapproximationimage (LL) as well as horizontal(HL), vertical (LH) and diagonal (HH)fordetail components.Inthe DWT decomposition input signal must be multiple of 2
n
. Where, n representsthe number of level.To analysis and synthesis of the original signalDWT provides the sufficient informationandrequireslesscomputation time.Watermarks are embeddedinthese regionsthat helptoincrease the robustness of thewatermark. Aone level DWTdecomposition processis shown inFigure 1.
Image
Figure1.One level DWTdecomposition process
2.2. Discrete Cosine Transform (DCT)
The DCT is the most popular transform function used in signal processing. It transforms a signalfrom spatial domain to frequency domain. Due to good performance, it has been usedin JPEGstandard for image compression. It is afunctionrepresents atechniqueapplied to image pixels in
LL
1
HL
1
LH
1
HH
1
L R
 
International Journal of Managing Public Sector Information and Communication Technologies (IJMPICT)Vol. 4, No. 2, June 2013
23
spatial domain in order to transform them into a frequency domain in which redundancy can bebranded.DCT techniques aremorerobust compared tospatial domain techniques.Suchalgorithms arerobust against simple image processing operationslikeadjustment,brightness,blurring,contrastandlow pass filteringand so on.But it is difficult to implementandcomputationally more expensive.Theone-dimensional DCT is useful in processing one-dimensional signals such as speech waveforms. For analysis of two-dimensional (2D) signalssuch as images, we need a 2D version of the DCT. The 2D DCT and 2D IDCT transforms isgiven by equation 1 and 2.Formulae of 2-D DCT:
(,)=( ) ( ) (,)cos(2 + 1)2(2 + 1)2(1)
Formulae of 2-D inverse DCT:
(,)=( ) ( ) (,)cos(2 + 1)2(2 + 1)2(2)
Where,
( ),( )=1, , = 02, , = 1 1
2.3. Singular Value decomposition (SVD)
The singular valuedecomposition(SVD)matrix is very useful in computer vision asadecomposition matrixand itis anefficient tool forimage transformations.The SVD ofa givenimage
in the form ofa matrixis defined as
= (3)
Where,S is the diagonal matrix that is
=
0. 000. 00.00.00....0.0
AndU and V are the orthogonal matrices
= =

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->