Fast Semi-blind Color Image Watermarking Scheme Using DWT and Extreme Learning Machine

Mishra, Anurag; Rajpal, Ankit; Bala, Rajni

doi:10.1007/978-3-319-62407-5_5

Cited by 6 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore protection of ownership rights in a scientific manner is of vital importance. One of the strong mechanism to deal with copyright protection of digital content is watermarking, which has undergone a thorough research over the past few decades and accordingly resulted into to development of numerous watermarking schemes [17,24,35,43,47,55,61,63,68,71,79]. However in view of the growing volume of data in every walk of life and its usage, it is imperative to explore new and more effective techniques to protect its ownership rights.…”

Section: Introductionmentioning

confidence: 99%

DWT based color image watermarking: a review

Giri

Quadri

Bashir

et al. 2020

Multimed Tools Appl

View full text Add to dashboard Cite

In today's knowledge driven society and technology driven economy, when data is considered to be one of the most important corporate resource of an organization, equally important is the security of the same to prevent unauthorized users to access and manipulate it. Watermarking being an appropriate solution for claiming authorization has prompted many researchers to work in this direction and accordingly a substantial amount of research work has been done using different schemes on various media types. The main impetus in writing this paper is to provide a broader view as how much work has been carried so far and what are the different dimensions that have been taken into consideration to watermark color images using discrete wavelet transformation. The paper will help prospective researchers to gain more insight about the work that has been done and pave a way for them to design and implement better techniques.

show abstract

Section: Introductionmentioning

confidence: 99%

DWT based color image watermarking: a review

Giri

Quadri

Bashir

et al. 2020

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Watermarkingofimagesisachallengingresearchareawhichinvolvesprinciplesandtechniquesof communications,signalprocessingandencryption (Cox,MillarandBloom,2002).Thisresearchhas manypotentialapplicationsandisdirectedtowardsdevisinganefficientwatermarkingschemewhich canbesuccessfullyimplementedinspatialand/ortransformdomains.Robustnessofembedding processisoneoftheimportantprerequisiteoftheseschemes.Evenifaschemeisprovedtoberobust enoughtotheselectedimageprocessingattacks,newattackscanbedevisedinfutureandthus,an apparentlyrobustwatermarkingschemecanbeprovedfatal.Thefinancialimplicationsofcopyright protectionareenormousandthereisarigorousincreaseofresearchactivitybyfamousresearch groupsacrosstheworldatpresent.Theresearchintheareaofimagewatermarkinginparticularhas takentwodirections.ThefirstoneisorientedtowardsitsSSPbasedindustrialapplicationslikethe onediscussedinthiswork.Thesecondorientationisevenmoreinterestingandabstractinnature.It majorlyinvolvestheuseofmachinelearningandothersoftcomputingtechniquestooptimallyembed thewatermarkintotheoptimallyidentifiedlocationswithintheimagesorvideos.Anumerousresearch activityiscarriedoutatpresentinthisdirection.Nearlyallkindsofsoftcomputingtechniques-ArtificialNeuralNetworks(ANNs) (Agarwal,2013(Agarwal, ,2015,FuzzyInferenceSystems(FIS) (Mishra, 2014(Mishra, ,2017,SupportVectorRegressionModels(SVMs) (Rajpal,2016)aswellasMetaheuristic techniques(GAanditsvariants)havebeenemployedforthispurpose.Inadditiontothesetechniques, hybridintelligentnetworksarealsoprofuselyusedtocarryoutwatermarkembeddingandextraction. Therefore,itissurethatsoonerorlaterthedevelopmentofnewandrobustwatermarkingschemes willofferacompetentandcomprehensivesolutiontotheproblemofintellectualpropertyrights(IPR) protectionandcontentauthentication.Toachievethisobjective,somestepsshouldbetaken.Such stepsmustconsistofacceptanceandimplementationofagloballyacceptedwatermarkingprotocol anddevisingwatermarkperformancebenchmarks.Tothisend,asasecondlineofresearchinsignal processingandparticularlytheimageprocessinginencrypteddomainhasattractedconsiderable attentionofresearchersinrecentyears (Erkin,Piva,Katzenbeisser,Lagendijk,Shokrollahi,Neven, andBarni,2007).Although,cryptosystemsplayasignificanttaskinsecuresignalprocessing,yet notallsystemsaresuitableforthispurpose.ThemainreasonisthatseveralcryptosystemslikeDES (DataEncryptionStandard)andAES(AdvancedEncryptionStandard)donotpreservealgebraic relationswiththeplaintextsaftertheencryptionprocess (Elgamal,1985).Ontheotherhand,special cryptosystemssuchasthosewhichofferhomomorphismkeepthisrelationship.Thus,theybecome suitableforsignalprocessingapplications.…”

Section: Introductionmentioning

confidence: 99%

Secure Image Watermarking in a Compressed SPIHT Domain Using Paillier Cryptosystem

Gupta

Mishra

Jain

2019

International Journal of Information System Modeling and Design

Self Cite

View full text Add to dashboard Cite

A secure solution to the problem of copyright infringement and content authentication is to carry out image watermarking in secure signal processing (SSP) domain. Homomorphic encryption is considered one such solution for image watermarking in this domain. The Paillier encryption is found to be suitable for image processing applications in general and for watermarking in particular. In this article, a detailed investigation is carried out by using Paillier cryptosystem for twelve different color images in a compressed domain. The compression of the host images is carried out by SPIHT (Set Partitioning in Hierarchical Trees) coding. The visual quality of the images post embedding and image processing attacks is assessed by using two full reference metrics, Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The performance evaluation of the Paillier cryptosystem vis-à-vis watermark application development is carried out by computing three benchmark metrics: number of pixels change rate (NPCR), unified average changing intensity (UACI) and encryption speed.

show abstract