Properties Of Digital Image Watermarking Computer Science Essay

Published Date: 02 Nov 2017

CHAPTER 2

LITERATURE REVIEW

This chapter reviews the appropriate background literature and describes the concept of digital image watermarking. The transmission of multimedia data became daily routine nowadays and it is necessary to find an efficient way to transmit through various networks. Copyright protection of multimedia data has become critical issue due to massive spreading of broadband networks, easy copying, and new developments in digital technology [18], [31]. As a solution to this problem, digital image watermarking became very popular nowadays.

The transmission of information takes place in different forms and is used in many applications. In a large number of these applications, it is desired that the communication to be done in secret. Such secret communication includes transfer of medical data, bank transfers, corporate communications, purchases using bank cards, a large percentage of everyday emails and etc.

N.Provos and P.Honeyman [26] says that steganography is different from cryptography and watermarking, although they all have overlapping usages in the information hiding process. Steganography security hides the knowledge that there is information in the cover medium, where cryptography reveals this knowledge but encodes the data as cipher-text. Similar to steganography, watermarking is about hiding information in other image, but difference is that watermark must be somewhat resilience against attempts to remove it. This technique of information hiding can be extended to copyright protection of multimedia content. Digital watermarking and steganography techniques are used to address digital right management, protect information, and conceal secrets. Information hiding techniques provide an interesting challenge for digital forensic investigations [23].

DIGITAL STEGANOGRAPHY

The word steganography comes from the Greek word Steganos, which means covered or protected, and – the word graphy, which means writing or drawing. Therefore, steganography means that literally, covered writing. Steganography is the technique of hiding information such that its presence cannot be detected and a communication is happening [27]. The advantage of steganography over cryptography is that messages do not attract attention to themselves. Therefore, whereas cryptography protects the content of a message, steganography can be said to protect both messages and communicating parties [23].

2.1.1 Properties of Steganography

All the steganographic algorithms need to fulfill the following basic requirements.

Invisibility- The invisibility of steganographic algorithm is the first and foremost requirement, since the steganography lies in its ability to be unnoticed by the human eye.

Payload Capacity- Steganography aims at hidden communication and therefore requires sufficient embedding capacity.

Robustness against Stastical Attacks- Statistical steganalysis is the practice of detecting hidden information through applying statistical tests on image data.

Independent of file format-The most powerful steganographic algorithms lies in the ability to embed information in any type of file.

2.1.2 Applications of Steganography

To have secure secret communication, where strong cryptography is not possible. In military applications, where even the knowledge that two parties communicate can be of large importance.

2.2 DIGITAL IMAGE WATERMARKING

Digital image watermarking is a kind of technology that embeds copyright information into multimedia content. An effective image watermarking mainly includes watermark generation, watermark embedding, watermark detection, and watermark attack [5], [1]. Digital image watermarking provides copyright protection to image by hiding appropriate information in original image to declare rightful ownership [12]. There are four essential factors those are commonly used to determine quality of watermarking scheme. They are robustness, imperceptibility, capacity, and blindness. Robustness is a measure of immunity of watermark against attempts to image modification and manipulation like compression, filtering, rotation, scaling, noise attacks, resizing, cropping etc. Imperceptibility is the quality that the cover image should not be destroyed by the presence of watermark. Capacity includes techniques that make it possible to embed majority of information. Extraction of watermark from watermarked image without the need of original image is referred to as blind watermarking. The non-blind watermarking technique requires that the original image to exist for detection and extraction. The semi-blind watermarking scheme requires the secrete key and watermark bit sequence for extraction. Another categorization of watermarks based on the embedded data is visible or invisible [6], [25]. Based on the robustness of the watermarks, watermarks are classified as robust watermarks, fragile watermarks and semi-fragile watermarks. Robust watermarks can resist malicious distortions, whereas fragile watermarks can easily destroyed by all image distortions and semi-fragile watermarks can be destroyed by certain type of distortions while resisting other minor changes.

The main applications of digital image watermarking include Digital Rights Management (DRM)/ Owner Identification, copyright protection and authentication. DRM can be defined as "the description, identification, trading, protecting, monitoring and tracking of all forms of usages over tangible and intangible assets [19]. It concerns the management of digital rights and enforcement of rights digitally. Copyright enables the identification of the copyright holder and thus protects rights in content distribution. It is used to prevent third party from copying or claiming the ownership of the multimedia content. Authentication in image watermarking refers to the integrity assurance of the image.

From application point of view, robust watermarks are suitable for copyright protection, because they can resist common image processing operations. On the other hand, fragile watermarks can be used to detect tampering and authenticate an image, because it is sensitive to changes. Semi-fragile watermarks are in some special cases of authentication and tamper detection.

According to the domain of watermark insertion, the watermarking techniques fall into two categories: spatial domain methods and transform domain methods. Many techniques have been proposed in the spatial domain such as LSB (Least Significant Bit) insertion method, the patch work method and the texture block coding method [6]. These techniques process the location and luminance of the image pixel directly. The LSB method has a major disadvantage that the least significant bits may be easily destroyed by lossy compression. Transform domain method based on special transformations, and process the coefficients in frequency domain to hide the data. Transform domain methods include Fast Fourier Transform (FFT),Discrete Cosine Transform(DCT), Discrete Wavelet Transform(DWT),Curvelet Transform(CT), Counterlet Transform(CLT) etc. In these methods the watermark is hidden in the high and middle frequency coefficients of the cover image. The low frequency coefficients are suppressed by filtering as noise, hence watermark is not inserted in low frequency coefficients [6]. The transform domain method is more robust than the spatial domain method against compression, filtering, rotation, cropping and noise attacks.

2.2.1 Properties of Digital Image Watermarking

The efficiency of a digital image watermarking process is evaluated according to the properties of perceptual transparency, robustness, computational cost, bit rate of data embedding process, false positive rate, recovery of data with or without access to the original image, the speed of imbedding and retrieval process, the ability of embedding and retrieval module to integrate into standard encoding and decoding process etc. [27-29].

To understand watermarking methods and determine their applications, one needs to know the properties of digital image watermarking.

Robustness- of a watermark refers to its ability to withstand non-malicious distortions. The watermarking scheme should be robust to any possible signal processing operations, as long as the quality of the watermarked data preserved.

Data Payload- is the encoded message size of a watermark in an image. On the other hand, multi-bit watermarks can carry textual or pictorial information [27].

Capacity- is the amount of information in an image. If multiple watermarks are embedded into an image, then the watermarking capacity of the image is the sum of all individual watermarks data payload [27].

While the robustness of the watermarking method increases, the capacity also increases where the imperceptibility decreases. There is a tradeoff between these requirements and this tradeoff should be taken into while the watermarking method is being proposed [19].

Imperceptibility – is the characteristic of hiding of the watermark so that it does not degrade the visual quality of an image. The imperceptibility of the watermark is tested by peak signal to noise ratio.

Fidelity- is the visual similarity between the watermarked image and its cover image.

Security- of the watermark is the ability of the watermark to resist malicious attacks. These attacks include intentional operations of another watermark insertion, modification, removal and estimation which aim at defeating the purpose of the watermarks.

Computational cost-is the measure of computing resources required to perform watermark embedding or detection processes. It can be measured using the processing time for a given computer configuration.

There are several ways of classifying watermarking methods. One of the widely adopted classifications is based on watermark robustness. Under this classification, watermark can be grouped into 3 types:

Robust watermarks are watermarks that can resist malicious distortions.

Fragile watermarks are easily destroyed by all image distortions.

Semi-fragile watermarks can be destroyed by certain types of distortions while resisting other minor changes.

Besides watermark robustness, watermarks can also be categorized into visible and invisible types. Visible watermarks are perceptible to a viewer. On the other hand, watermarks are imperceptible and do not change the visual appearance of the images.

Depending upon the application, the properties, which are used mainly in the evaluation process, vary. For example, in the video indexing application, evaluating the robustness of a watermarking scheme to any signal processing is meaningless, since there is no case that the video passes through some signal processing operation. In the covert communication application, it is better to use a watermarking scheme that does not need the original data during the watermark detection process, if real television broadcasting is used as the communication channel, while most of the watermarking schemes in other applications need the original data during the detection process. If the application is the copyright protection, the other owner of the original data may wait for several days to insert or detect watermark, if the data is valuable for the owner. On the other hand, in a broadcast monitoring application, the speed of the watermark detection algorithm should be as fast as the speed of real time broadcasting. As a result, each watermarking application has its own requirements and the efficiency of the watermarking scheme should be evaluated according to these requirements [30].

The owner of the original data wants to prove his ownership in case of original data is copied, edited and used without permission of the owner. In the watermarking research world, this problem has been analyzed in a more detailed manner [19].

2.2.2 Applications of Digital Image Watermarks

Digital image watermarking techniques have been proposed to be implemented in many applications. Some major groups of its applications are:

Digital Rights Management(DRM)/Owner identification

DRM can be defined as the description, identification, trading, protecting, monitoring and tacking of all forms of usages over tangible and intangible assets. It concerns the management of digital rights and enforcement of rights digitally.

Copyright protection

It enables the identification of the copyright holder and thus protects the rights in content distribution. It is used to prevent third parties from copying or claiming the ownership of the digital media. Robust watermarks are embedded into an image to protect the rights of the owners. It should be possible to detect the watermark despite common image processing, geometrical distortions, image compression, and many other image manipulations. The successful detection of the watermark can positively identify the owner.

Authentication

Authentication in image watermarking refers to the integrity assurance of the image. The applications related to image authentication are the validation of cultural heritage paintings, medical records and digital artworks.

Other Applications:

There are many other applications where digital image watermarking methods have been proposed as a technology enabling tool. Some of them are:

Broadcast monitoring- watermarks embedded into advertisement sections of broadcast. It is used to track the broadcast of a particular file over a channel.

Device control- watermarks embedded into radio and television signals can be used to control features of a receiver.

Medical Applications- used in X-ray film references where they are marked with a unique ID of the patient.

Fingerprinting- to convey information about the recipient of the digital media.

Copy control- watermarks detected in a video content are used to control the functionality of a watermark complaint recorder.

Application wise robust watermarks are suitable for copyright protection, because they can resist common image processing operations. On the other hand, fragile watermarks can be used to detect tampering and authenticate an image, because it is sensitive to changes. Semi-fragile watermarks are usually applied in some special cases of authentication and tamper detection. These cases may consider lossy image compression as legitimate changes while highlighting geometrical distortions as intentional attacks.

2.2.3 Key differences between watermarking and Steganography

Digital Image Watermarking

Inserts information related to either to host signal or its owner.

Main goals are copyright protection and information authentication.

It is either visible or imperceptible.

It is for communications point-to-multiple points.

Capacity is not an important issue

Robustness is an important issue

Digital Steganography

Must not only be imperceptible but also statistically undectable.

Is for point-to-point communications.

Main goal is covert communication.

Inserts any kind of information.

Capacity is an important issue.

May or may not be robust.

2.3 DIGITAL IMAGE WATERMARKING ALGORITHMS

Digital image watermarking algorithms are classified into three categories namely spatial domain methods, feature domain methods and transform domain methods. In spatial domain methods, the watermark is embedded directly into pixel values of the original image. In feature domain methods, the watermark embedding depends upon the region, boundary and object characteristics. In transform domain methods, the watermark is embedded into the transformed coefficients of the original image. The transform methods have been found to have greater robustness, when the watermarked images tested after having been affected by different attacks.

2.3.1 Spatial Domain Techniques

Many spatial techniques are based on adding fixed amplitude pseudo noise sequences to an image. Pseudo random noise (PN) sequences are used as the spreading key when considering the host media as the noise in spread spectrum system, where the watermark is the transmitted multimedia content. Many techniques have been proposed in the spatial domain such as LSB (Least Significant Bit) insertion method, the patch work method and the texture block coding method [6]. These techniques process the location and luminance of the image pixel directly. The LSB method has a major disadvantage that the least significant bits may be easily destroyed by lossy compression. The invisibility of the watermark is achieved on the assumption that the LSB data are visually insignificant. In general, the techniques that modify the LSB of the data using a fixed magnitude PN sequence are extremely sensitive to signal processing operations and weak to watermark attacks. The contributing factor to this weakness is the fact that the watermark must be invisible. As a result, the magnitude of the embedded noise is limited by the smooth regions of the image, which most easily exhibit the embedded noise.

2.3.2 Transform Domain Techniques

Transform domain method based on special transformations, and process the coefficients in frequency domain to hide the data. Transform domain methods include Fast Fourier Transform(FFT), Discrete Cosine Transform(DCT), Discrete Wavelet Transform(DWT), Curvelet Transform(CT), Counterlet Transform(CLT) etc. In these methods the watermark is hidden in the high and middle frequency coefficients of the cover image. The low frequency coefficients are suppressed by filtering as noise, hence watermark is not inserted in low frequency coefficients [7]. The transform domain method is more robust than the spatial domain method against compression, filtering, rotation, cropping and noise attack etc.

Many transform based digital image watermarking techniques have been proposed by researchers and scientists. To embed a watermark, first transform is applied on the cover image and then modifications are made to the transformed coefficients.

Cox et al [32] find parallels between spread-spectrum communications and watermarking and used a frequency domain transform to convert an image into another domain.

In frequency domain, a sequence of values I0= I0[1], I0[2], ……I0[n] are extracted from the given carrier image and then this sequence is modified as per the requirement. The watermark is a sequence real numbers w = w[1], w[2] ...…w[n]. Each value of this watermark sequence is chosen independently from the Gaussian distribution with zero mean and with variance unity.

Three different formulas to embed watermark, whose difference lies in their embedding characteristics and in their invertibility are given below:

Iw[i] = I[i] +αw[i] ………………………… (2.1)

Iw[i] = I[i] (1+αw[i]) ……………………… (2.2)

Iw[i] = I[i] +exp (αw[i]) ……………………. (2.3)

Where α is the scaling or watermark strength parameter, which influences he robustness as well as the imperceptibility of the watermarked image.

Watermarking can be implemented in frequency domain such as proposed by Cox et al [32], where the embedding technique is based on DCT and Pseudo Noise sequence. The watermark extraction is based on the knowledge of cover image and the frequency locations of the watermark. The normalized correlation coefficient is computed and set to a certain threshold. If the normalized correlation coefficient is large enough, the watermark is detected. This Cox et al method is robust to image scaling, JPEG compression, dithering, cropping, and rescanning.

Another watermarking scheme in frequency domain is wavelet transform technique. Barni et al [33], proposed a watermarking method on decomposition of wavelet transforms. The technique based on the decomposition of input cover image into low and high frequency components with different orientations. A Discrete Wavelet Transform is applied to the cover image. The watermark is inserted into the highest level subbands as per the following rule:

IwLH [i,j]= I0LH[i,j]+αβLH[i,j]w[iN+j] ………..(2.4)

IwHH [i,j]= I0HH[i,j]+αβHH[i,j]w[MN+iN+j] ……(2.5)

IwHH [i,j]= I0HH[i,j]+αβHH[i,j]w[2MN+iN+j] ……….(2.6)

Where α is the global parameter for the watermarking strength, β is the local weighting factor and w is the pseudo random binary sequence. The masking characteristics of human visual system depend on this local weighting factor. The correlation between the watermark DWT coefficients and the watermark sequence is computed to retrieve the watermark.

The similarity between the correlation method and Barni’s method is shown by Cox et al. [32]. This algorithm is formulated as a correlation by defining the pattern with the same dimensions as that of coefficient matrix. The pattern values are determined by the influence of the corresponding frequency coefficients. The pattern is zero for coefficients not considered in the evaluation. The pattern for the pair coefficients is either 1 or -1. Thus, the sign of the correlation directly depends on the relation of the pair of coefficients.

Fractal watermarking schemes are based on fractal compression, which is developed based on iterated function systems. The fractal encoding algorithm partitions the original image into non- overlapping domain cells. The image is covered with overlapping domain cells. For each range cell, the corresponding domain cell and transform are searched to determine the best cell range. This step is computationally expensive. The range of transforms typically includes affine transforms, change of brightness and contrast. This transform describes the self-similarity between range cell and the corresponding domain cell. To embed the watermark, the range cells are restricted by the encoded information. To retrieve the watermark from the watermarked block, the corresponding domain cells reveal the embedding information.

Samesh Oueslati et al, [13] proposed an adaptive image watermarking scheme based on Multi-Layer Feed forward (MLF) neural networks. In this method the host image is first decomposed into non-overlapping 8x8 blocks, and the DCT process is performed for each block. Coefficients are then selected for watermark insertion. Human Visual System (HVS) is adopted to further ensure the watermark invisibility. Then the luminance sensitivity, frequency sensitivity, texture sensitivity and entropy sensitivity are computed and used to as the inputs of the NNS. In this paper, neural networks are used to automatically control and maximum image–adaptive strength watermark.

Cheng-Ri Piao et al, [16] proposed a blind watermarking algorithm based on HVS and RBF neural network for digital images. In this method, RBF is implemented while embedding and extracting watermark. The human visual system model is used to determine the watermark insertion strength. The inserted watermark is a random sequence. The secret key determines the beginning position of the image where the watermark is embedded. This process prevents possible pirates from removing the watermark easily.

Nizar Sakr et al, [20] proposed an adaptive wavelet-based watermarking algorithm that is based on the model of a HVS and a Fuzzy Inference System (FIS).In this method; Sugeno-type fuzzy model is exploited in order to determine a valid approximation of the quantization step of each DWT coefficient. Furthermore, the HVS properties are modeled using biorthogonal wavelets to improve watermark robustness and imperceptibility.

Wu Bo XiaoMing Cui Chao Zhang [21] developed digital image watermarking encryption algorithm using fractional Fourier transform, which is robust against JPEG compression and Gaussian low-pass filtering.

Alain Tremeau and Damien Muslet [38] explained in detail about recent trends in color image watermarking. Teruya Minamoto and Kentaro Aoki [39] proposed a blind digital image watermarking method using interval wavelet decomposition.

FENG Yang, LUO Senlin, PAN Limin [40] proposed an extensive method to detect the image digital watermarking based on the known template. This method is used to extract some special features from DWT, DCT and spatial domains of the template and image. Then these features are used to detect the watermark.

Yu Chang-hui, FENG Wan-li and Zhou Hong [41] proposed the digital image watermarking technology based on neural networks. In this method they proposed three stage watermarking technique to improve the robustness of the watermarked image.

Jing-Jing Jiang and Chi-man Pun [42] proposed digital image watermarking based on patchwork and radial basis neural network. In this method two special subsets of the cover image features are selected embed watermark. One subset is used to add a small constant while the other is used to subtract the same from other patch.

Xinhong Zhang and Fan Zhang [43] proposed a blind watermarking algorithm based on neural network. In this method Hopfield Network and the Noise Visibility Function are used for adaptive watermark embedding.

Quan Liu and Xumei Jiang [44] proposed design and realization of a meaningful digital watermarking algorithm based on RBF neural network. In this method the radial basis function network and discrete cosine transform are used to simulate human visual specialty to determine the intensity of watermark embedding.

Chuan-Yu Chang, Wen-Chih Shen and Hung-Jen Wang [45] proposed robust digital audio watermarking in DWT domain using counter-propagation neural network. In this method the db4 filter of the Daubechies wavelet is applied to decompose the coefficients of the host image to improve the robustness.

Ju-Liu, Xingang, Montse Najar and Miguel Angel lagunas [46] proposed robust digital watermarking scheme based on ICA. In this method the combination of discrete cosine transform and independent component analysis is applied to improve the robustness of the watermarked image.

Cong Jin, Shu-Wei Jin [47] proposed an adaptive digital image watermark scheme based on fuzzy neural network for copyright protection. In their method fuzzy set theory is adopted to get rid of the slow training speed network parameter sensibility.

Fan Zhang and Hongbin Zhang [48] proposed different applications of neural network to improve the watermarking capacity. In this method a blind watermarking based on Hopfield network is proposed to improve the robustness of the watermarked image.

Ahmad R Naghsh Nilchi, Ayoub Taheri [49] proposed a new robust digital image watermarking technique based on the discrete cosine transform and neural network. In this method Full Counter Propagation Neural Network is implemented to simulate the visual and perceptual characteristics of the host image.

Qun- ting Yang and Tie-gang Gao, Li Fan [50] proposed a novel robust watermarking scheme based on neural network. In their method three identical watermarks are embedded into the low frequency subbands of the cover image to improve the performance of the watermarked image in terms of robustness and imperceptibility.

Song Huang and Wei Zhang et al [51] proposed a blind watermarking scheme based on neural network. This watermarking technique hides a scrambled watermark into an image and utilizes HVS characteristics during the embedding process.

Santi P. Maity and Seba Maity [52] proposed multistage spread spectrum watermark detection technique using fuzzy logic. In this method they proposed a new model of watermarking using spread spectrum to reduce bit error rate at the expense of computational complexity.

Mukesh C. Motwani, Rakhi C. Motwani et al [53], proposed wavelet based fuzzy perceptual mask for images. In this method they developed non-linear HVS model for perceptual masking with brightness, and sensitivity and texture as input variables to fuzzy system.

Gursharajeet Singh Karla et al [54] proposed robust blind digital image watermarking using DWT and dual encryption technique. They developed an algorithm based on properties of random sequence generated by Chaos and Arnold transformations for robust digital image watermarking.

Mukesh Motwani, Nikhil Beke et al [55] developed an adaptive fuzzy watermarking for 3D Models. In this method they developed an algorithm based on wavelet and fuzzy logic for 3D models, which is robust to smoothing, cropping, affine operations and noise attacks.

Mengyue Hu, Xiaolin Tian and Shaowei Xia [56] developed robust digital image zero-watermarking algorithm based on CDMA Technology, which has good performance under multiple attacks.

Pankaj U Lande, Sanjay N. Talbar and G.N Shinde [57], proposed an image adaptive watermarking using fuzzy logic. In this they developed real time low cost and robust watermarking hardware based on FPGA.

Soheila Kiani and Mohsen Ebrahimi Moghaddam [58] presented fractal based digital image watermarking using fuzzy C-Mean clustering, which is robust against JPEG compression, median filtering and additive noise.

Said E. El-Khamy et al [59] proposed a new perceptual image watermarking technique based on adaptive fuzzy clustering. In this method first the host image is decomposed into DCT blocks, then classified using adaptive fuzzy classification and perceptually embedded into each block to increase robustness against attacks.

Jianzhen Wu and Jianying Xie [60] developed an adaptive image watermarking scheme based on HVS and fuzzy clustering theory to embed watermark into the larger coefficients in mid-band frequency to improve the robustness of the watermarked image.

Reza Mortezaei et al [61] proposed a new lossless watermarking scheme based on fuzzy integral and DCT domain, which is robust against attacks such as compression, filtering, cropping and other noise attacks.

LI Li Zong and Gao Tie gang [62] proposed a zero-watermarking algorithm based on fuzzy adaptive resonance theory with three roles in the algorithm namely the signer, the verifier and trusted authority.

Nizar Sakr, Jiying Zhao and Voicu Groza [63] presented an adaptive image watermarking based on a dynamic fuzzy inference system. This algorithm utilizes HVS model to adjust and select the appropriate watermark strength to improve the robustness.

Hung-Jen Wang, Chuan-Yu Chang et al [64] developed a DWT-based robust watermarking scheme with fuzzy ART to protect the intellectual property rights of digital multimedia. The proposed method is robust to internal attacks, geometric distortions and image processing attacks.

Chip-Hong Chang,Zhi Ye and Mingyan Zhang [65] proposed fuzzy-ART based digital watermarking scheme to insert a visually recognizable image and a weighted recovery technique is developed to improve the robustness of the watermarked image against normal lossy compression attacks.

Ming-Shing Hsieh [66] developed an image watermarking based on fuzzy inference filter to embed signatures in images to attest the owner identification, to discourage unauthorized copying, to provide transparency and robustness.

Hai-Yan Tu Jiu-Lun Fan et al [67] presented a robust watermark algorithm based on Ridge let transform and fuzzy C-means to obtain a sparse representation for straight edge singularity. In this method, FCM clustering is applied to classify the image pieces into frat regions and texture regions adaptively.

Prof.Sharvari C.Tamane, and Dr.R.R. Manza [68] proposed 3D Models watermarking using fuzzy logic based on the human visual system in wavelet domain. This method gives a perceptual value for each corresponding wavelet coefficient to improve the robustness of the watermarked image.

Hsiang-Cheh Huang, Yueh-Hong Chen and Guan-Yu Lin [69] developed a fuzzy-based bacterial foraging algorithm to design an effective fitness function to improve the quality and robustness of the watermarked image.

Lei Li, Wen-Yan Ding and Jin-Yan Li [70] discovered a novel robustness image watermarking scheme based on fuzzy support vector machine to select an appropriate degree of membership to reflect the importance of the sampling points. In this method, the image sub-block is divided into categories namely weak texture and strong texture and watermark is embedded into strong structure blocks to improve the robustness.

Jun Fan, Yiquan Wu [71] developed a watermarking algorithm based on kernel fuzzy clustering and singular value decomposition in the complex wavelet transform domain where image low- frequency background and high frequency texture features are used as fuzzy clustering vectors to determine the different embedding strength.

Nizar Sakr, Jiying Zhao and Voicu Groza [72] proposed a dynamic fuzzy logic approach to adaptive HVS-based watermarking to adjust and select the appropriate watermark length to provide a more robust and imperceptible watermark.

Hajime Nobuhara, Witold Pedrycz, and Kaoru Hirota [73] presented digital watermarking algorithm using image compression method based on relational equation to improve the imperceptibility of the watermarked image. In this method, image compression and reconstruction is done on 100 images and confirmed that the signed image is distinguishable from the unsigned image.

Glumov N.I. Mitekin V.A [74] developed a new block wise algorithm for large scale images. In this method, the host image is divided into non-overlapping fragments and the average centered magnitude spectrum is calculated for the entire host image to provide better robustness.

Farooq Husain, Ekram Khan and Omar Farooq [75] proposed DFRFT-domain digital image watermarking. In this method randomly distributed sequence is used as a watermark to modify discrete fractional Fourier transform coefficients of the cover image.

Jindong Xu, Huimin Pang, Jianping Zhao [76], developed digital image watermarking algorithm based on fast curvelet transform. In this method the carrier image is decomposed by fast curvelet transform and watermarked image is scrambled by Arnold transform.

Mahasweta J.Joshi et al [77] proposed digital image watermarking in DCT-DWT domain to protect watermarked images from illegal manipulations. This algorithm is robust against white noise, Gaussian filtering and sharpening filter attacks.

J.Anitha and S.Immanuel Alex Pandian [78] developed color image digital watermarking scheme using SOFM based on codebook partition technique to embed watermark bit sequence into the vector quantization encoded blocks, which is robust against compression.

G.Thirugnanam and S.Arulselvi [79] implemented wavelet packet based robust digital image watermarking and extraction using independent component analysis with high peak signal to noise ratio compared wavelet transforms.

Hanjie Ji, Jie Zhu and Hongqing zhu [80] developed combined blur and RST invariant digital image watermarking using complex moment invariants to protect the watermarked images against geometric image manipulations with good robustness.

Mohammadreza Ghaderpanah and A.amza [81] presented nonnegative matrix factorization scheme for digital image watermarking to improve the performance of the data embedding system and resist a variety of intentional attacks and normal visual processes.

Ming-Shing Hsieh, Din-Chang Tseng and Yong-Huai Huang [82] proposed a technique to hide digital watermarks using Multiresolution wavelet transform. In this method the watermark is detected by comparing an experimental threshold with the extracted values. They also proposed multi energy watermarking scheme based on qualified significant wavelet tree to improve the robustness of the watermarked images.

E. Chrysochos, V.Fotopoulos and A.N. Skodras [83] developed robust digital image watermarking based on chaotic mapping and discrete cosine transform to protect the watermarked images against noise addition, filtering, JPEG compression and geometric manipulations.

Saeed K. Amiirgholipour, Ahmad R. Naghsh-Nilchi [84] proposed robust digital image watermarking based on Joint DWT-DCT technique to provide higher robustness noise attacks and enhancement.

Houng-Jyh Mike Wang, Po-Chyi Su and et al [85] developed wavelet- based digital image blind watermarking which is robust against signal processing attacks and compression. In this method blind watermark retrieval technique is used to detect the embedded watermark without the need of the original image.

Bum-Soo Kim et al [86] proposed robust digital image watermarking method against geometrical attacks by improving Fourier Mellin transform based watermarking. This method modifies and reorders function blocks of Fourier Mellin transform by the use of an invariant centroid as the origin.

Kilari Veeraswamy, B.Chandra Mohan et al [87] developed an image compression and watermarking scheme using scalar quantization and counterlet transform with double filter bank structure based on the Laplacian Pyramid. This method is superior to wavelet transform method when the image contains more contours and is robust to normal image attacks.

B.chandra Mohan and S.Srinivas Kumar [88] implemented the robust digital image watermarking scheme using counterlet transform with multiple descriptions and quantization index modulation.

B.chandra Mohan, S.Srinivas Kumar and B.N.Chatterji [89] developed a robust digital image watermarking scheme using singular value decomposition, dither quantization and edge detection which is resilience to image attacks.

Chandra Mohan.B and Srinivas Kumar.S [90] proposed robust multiple image watermarking scheme using discrete cosine transform with multiple descriptions, which is robust to local and global attacks.

Sanjeev kumar, Blasubramanian Raman and Manoj Thakur [91] implemented real coded genetic algorithm based stereo image watermarking in discrete wavelet transform domain. In this method a pair of stereo images is used to generate a disparity-image watermark to embed into the degraded cover image by modifying singular values.

Fouad Khelifi and Jianmin Jiang [92] developed perceptual image hashing based on virtual watermark detection to provide better robustness against geometric attacks and image processing manipulations.

Pankaj U.Lande, Sanjay N.Talbar and G.N.Shinde [93] developed hardware for FPGA prototype of robust watermarking JPEG 2000 encoder, which is robust against scaling, rotation and most of the geometric attacks.

Pankaj.U.lande, Sanjay N.Talbar and G.N.Shinde [94] proposed fuzzy logic approach to encrypt watermark for still images in wavelet domain based on FPGA. This watermarking system is implemented by hardware to meet real time constraints related to robustness and imperceptibility.

Hanaa A.Abdallah, et al [95] developed blind wavelet-based image watermarking based on inserting the watermark bits into the coarsest scale wavelet coefficients by performing three-level wavelet decomposition.

Gaurav BHatnagar et al [96] proposed DWT-SVD based dual watermarking scheme to improve the protection and robustness by embedding dual watermarks into the cover image. In this method the secondary watermark is easily detected but the primary watermark is severely distorted.

Hamed Modaghegh et al [97] developed a new adjustable blind watermarking based on GA and SVD considering image complexity and robustness. This algorithm is an adjustable solution by changing the fitness function so that watermarking technique can be converted into robust, fragile or semi-fragile types.

Mohammad Reza Soheili [98] presented a blind wavelet based logo watermarking to resist cropping. In this method a binary logo is embedded into LL2 subband of the cover image using quantization. The robustness of the algorithm is increased by adding two dimensional parity bits to the binary logo.

Juan R. Hernandez, et al [99] proposed watermarking techniques in DCT-domain for still images. In this method spread spectrum technique is implemented in DCT domain to increase the robustness and imperceptibility of the watermarked image.

Samira Mabtoul, Elhassan Ibn Elhaj, and Aboutajdine [100] developed robust semi-blind digital image watermarking technique in DT-CWT domain to increase the security of the watermarked image. In this method two chaotic maps are generated, one is used to determine the blocks of the cover image to embed watermark, while the other is used to encrypt the watermarked image.

S.Saryazdi, H.Nezamabadi-pour, and A.Hakimi [101] proposed a blind watermarking scheme for binary image authentication to detect any alterations. In this method the binary host image is divided into 2x2 sub-blocks and the last pixel is predicted from its neighbors.

Xiang-Wei Zhu [102] developed blind watermark detection algorithm based on generalized Gaussian distribution to protect copyright, intellectual and material rights of distributors, authors and buyers. In this method a blind watermark detection technique is developed according to the method of maximum likelihood estimation and the algorithm is very much effective against most of the image attacks.

Pik Wah chan, Michael R.Lyu, and Ronald T.chin [103] proposed a novel scheme for hybrid digital video watermarking based on scene change analysis and error correction code, which is robust against the attacks such as frame dropping, stastical analysis and averaging.

Chih-Wei tang and Hsueh- Ming Hang [104] developed feature-based robust digital image watermarking scheme using image normalization and Mexican Hat wavelet interaction. This scheme can survive low quality JPEG compression, sharpening, median filtering, color reduction, cropping and rotation attacks.

Chin-Chen Chang and Henry Chou [105] implemented a new public-key oblivious fragile watermarking for image authentication using discrete cosine transform to improve the vulnerability towards different image attacks.

Sudip ghosh, Pranab Ray et al [106] proposed spread spectrum image watermarking with digital design for greater robustness. In this method Field Programmable Gate Array has been developed using VLSI and the circuit is integrated into the existing digital still camera frame work.