Documents

A Robust DWT Digital Image Watermarking Technique Basis On Scaling Factor

Description
At the present time, the aptitude in contemplation of accessing as well as sharing images has become progressively facile with the Internet allowing people to procure information remotely from anywhere in the entire world. Moreover, there has been
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  International Journal of Computer Science, Engineering and Applications (IJCSEA) Vol.2, No.4, August 2012 DOI : 10.5121/ijcsea.2012.240763  A Robust DWT Digital Image WatermarkingTechnique Basis On Scaling Factor Heena Shaikh,Mohd.Imran Khan, YashovardhanKelkar Department of Information Technology, MIT, Ujjain heenushaikh@gmail.com , imran.khan@mitmandsaur.info, yasho028@rediffmail.com  Abstract:  At the present time, the aptitude in contemplation of accessing as well as sharingimages has become progressively facile with the Internet allowing people to procure information remotely from anywhere inthe entire world. Moreover, there has been also an expansion with regard to the number of the still digitalimages over the internet for the sake of the fact that a vast number of millions of people are capturingdigital photos.Fundamentally, the procedure of digital watermarking can be delineated as a method for embedding information into another signal (a digital signal). In case of digital images, the embedded information can be either visible or hidden from the user. In this project, we will concentrate onimperceptible watermarks. The principal intention of digital watermarks is to provide copyright  protection for intellectual property that is in digital format. Typical usage scenarios for watermarking aresuch as copyright protection and data authentication.In this paper, we describe an imperceptible and arobust DWT digital Image Watermarking algorithm.The algorithm watermarks a given digital image usinga Discrete Wavelet Transform.  Keywords:  Digital watermarking,DiscreteWaveletTransform(DWT),Peak Signal to Noise Ratio,ScalingFactor,GaussianNoise,Salt and Pepper. I.Introduction The DWT transform,Wavelets are special functions which, in a form analogous to sines andcosines in Fourier analysis, are used as basal functionsfor representing signals. For2-D images,applying DWT corresponds to processing the image by 2-D filters in each dimension. Thefilters divide the input image into four non-overlapping multi-resolution sub-bands (LL 1 ), (LH 1 ),(HL 1 ) and (HH 1 ). The sub-band (LL 1 ) represents the coarse-scale DWT coefficients while thesub-bands (LH 1 ), (HL 1 ) and (HH 1 ) represent the finescale of DWT coefficients[1]. To obtain thenext coarser scale of wavelet coefficients, the sub-band (LL 1 ) is further processed until some final scale “ N ”is reached. When “ N ” is reached we will have 3N+1 sub-bands consisting of the multi-resolution sub-bands (LL N ) and (LH X ),(HLx) and (HHx ) where “X”ranges from 1 until “ N ”[1][2] .Due to its excellent spatio-frequencylocalization properties, the FDWT is very suitable toidentify the areas in the host image. where a watermark can be embedded effectively. In  International Journal of Computer Science, Engineering and Applications (IJCSEA) Vol.2,No.4, August 2012 64 particular, this property allows the exploitation of the masking effect of the human visualsystem such that if a DWT coefficient is modified, only the region corresponding to thatcoefficient will be modified. In general most of the image energy is concentrated at the lowerfrequency sub-bands (LL X ) and therefore embedding watermarks in these sub bands may degradethe image significantly[3]. Yes, NoOriginal It’s True It’s fakeDigital Image Fig. I Digital image Watermarking The embedding and detecting procedure for watermarking techniquebased on DWTtransform.[4]Computing PSNR function (peak single-to-noise ratio)based on scaling factor oftheresultant watermarked images from the techniques DWT for the purpose of measuring thedistinctive distortion between the cover image and the watermarked image.[5]Applying thecheckmark software by means ofMSEfunction for the srcinal watermarks and extractedwatermarks from the DWTtechnique.[5] II.Proposed Watermarking Scheme In the case of one dimensional signal, the signal itself is to be divided into two groupsof frequency components as low frequency components and high frequency components which aremainly determined as the 1 st pass of the low-pass and high-pass frequencies, While the high-bandfrequency group would remain unchanged, the low-band frequency group will be thendivided up into two other inner groups of frequencies causing the 2 nd pass of the low-pass andhigh-pass frequencies. The same process is to be continued in such an arbitrary number of timesmaking the next passes by dividingthe low-band frequency blocks. The decomposition processcan be simply constructed by applying the previous technique into the one-dimensional signalX(N)[6].However, the srcinal signal  X(N) can be also reconstructed by using the same frequencycoefficients which have been used through the decomposition process of the DWT. Thereconstruction process is determined as the inverse DWT (IDWT).[6][7]The function of the DWT transform domain is based on the watermarking technology iscomposed oftwo major parts which are the encoding and decoding procedures. Those twoprocedures are plainly the key for implementing the embedding as well as the extracting systemby way of the DWT transform.[7]. Watermark EmbeddingTransmissionWatermark  Extraction Compare WatermarkAttackWatermark   International Journal of Computer S Fig. II II.I Embedding Techni This technique will decompose tbands through the first pass afrequency bands where it has thinto a 2 nd level (pass).[6]Secwatermark signature into the rfrequency coefficients without must add the signal of the banGaussian Noise and modifying tmoderating the srcinal signal image would be performed appr CovFi LL2HL2LH2HH2 HL1LH1HH1  ience, Engineering and Applications (IJCSEA) Vol.2,No.4  Sketch Map of Image DWT Decomposed ue for DWT  he cover image of the two dimensional DWT into f  s (  LL 1 ), (  LH  1 ), (  HL 1 ) and(  HH  1 ) frequency coef  lowest resolution of the 1 st pass (LL1)can be als   ondly, we are to apply the Gaussian Noise and set of the available frequency bands which incl dealing with (  LL ) regions from all over the passes ds where the large frequency components to the em without hich resides in the (  LL ) band; thereafter, the priately.[5] r ImageTest ImageWatermarked Imageg. IIIEmbedding Technique for DWT LL2HL2LH2HH2 HL1LH1HH1 , August 2012 65  ur frequencyficients. The decomposed an insert theude the high (levels). We signal of the watermarked LL2HL2LH2HH2 HL1LH1HH1  International Journal of Computer Science, Engineering and Applications (IJCSEA) Vol.2,No.4, August 2012 66 II. ExtractionTechnique for DWT Incontemplationofachieving this procedure, we should have the cover image and thewatermarked imagereadily applicable.Consequently,the DWT decoding technique willdecompose those two images into four frequency bands through the 1st pass as describedpreviously.Afterward, we aretoselectfor instanceoneofthosebands where the largefrequencies reside through one ofthe levels (passes) in the decomposed cover image and the decomposed watermark. Let’s suppose the selected band from both decomposed images is (HH1), we have then to compare the difference of the frequency coefficients in those bands of thedecomposedimages and examine their cross correlation. Subsequently, if the cross correlationhas detected a peak, then the watermark signature will be extracted; if not, then the sameoperation will continue on comparing the rest of the other bands consist the highfrequencycomponents from both of the decomposed images and investigate their cross correlation until thepeak is detected; correspondingly, the watermark signature will be latterly recovered.[4] Watermarked ImageRetrieved Image(No Attack)Watermarked ImageRetrieved Image(Gaussian Noise)Watermarked ImageRetrieved Image(Salt and Pepper)Fig. IVExtraction Technique for DWT  International Journal of Computer S III. Performance Eval For testing the performance of MATLAB. Inthe following exused as host image to embedwthe watermark. In order to test signal attacks, which includes image cutting and rotation. The watermark under different kind are shown in Tab. I. Simulatiagainst many common different compression, Gaussian noise. (b), (  ience, Engineering and Applications (IJCSEA) Vol.2,No.4 ation  this algorithm, theexperiments is simulated with eriments, the gray-level image with sizeof384*termark. Another binary image with size of196*21 the performance, the watermarked image suffers s filer, sharp enhancing,adding salt noise, image simulation results, including the watermarked imag of signal attack [9], The exact PSNR values of thn results suggest that this watermark algorithm c types of attacks such as adding salt and pepper (a)(b) (c)(d)(a) Original Image c), (d) Watermarked Image with differentScaling FactorFig. V Attack-Gaussian Noise , August 2012 67  the software84 “Lena” is 0 “ key ” is as  me different compression, and distillede processings an be robust noise, image
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks