SVM regression based robust image watermarking technique in joint DWT-DCT domain
01 Jul 2017-pp 1426-1433
TL;DR: A robust image watermarking technique has been proposed based on the combination of discrete wavelet transform (DWT) and discrete cosinetransform (DCT) and the experimental results show that the algorithm provides adequate robustness against both the geometric and non-geometric attacks.
Abstract: In this paper, a robust image watermarking technique has been proposed based on the combination of discrete wavelet transform (DWT) and discrete cosine transform (DCT). SVM regression model has been incorporated for geometric distortion correction to achieve improved robustness against de-synchronization attacks such as rotation, translation etc. Low order Pseudo Zernike (PZ) moments have used as a feature vector in SVM regression model. The 3-level DWT transformed DCT coefficients are modified to embed a binary bit. The performance of algorithm has been observed against both intentional and non-intentional attacks. The scheme provides an average imperceptibility of around 42.45 dB. The effect of rotation and translation attack has been estimated using trained SVM regression model. The robustness against de-synchronization attack is performed after correcting the attacked watermarked images. The experimental results show that the algorithm provides adequate robustness against both the geometric and non-geometric attacks.
Citations
More filters
TL;DR: A broader view is provided 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.
Abstract: 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.
12 citations
TL;DR: In this paper , a convolutional stacked denoising autoencoder (CSDAE) is utilized for producing hash codes that are robust against different content preserving operations (CPOs).
Abstract: With the advancement of technology, it has become easier to modify or tamper with digital data effortlessly. In recent times, the image hashing algorithm has gained popularity for image authentication applications. In this paper, a convolutional stacked denoising autoencoder (CSDAE) is utilized for producing hash codes that are robust against different content preserving operations (CPOs). The CSDAE algorithm comprises mapping high-dimensional input data into hash codes while maintaining their semantic similarities. This implies that the images having similar content should have similar hash codes. To demonstrate the effectiveness of the model, the correlation between hash codes of semantically similar images has been evaluated. Subsequently, tampered localization is done by comparing the decoder output of the manipulated image with the hash of the actual image. Then, the localization ability of the model was measured by computing the
f
1
scores between the predicted region and the original tampered region. Based on the comparative performance and receiver-operating characteristics (ROC) curve, we may conclude that the proposed hashing proposed algorithm provides improved performance compared to various state-of-the-art techniques.
7 citations
TL;DR: The experimental results show that the watermarked test images are highly transparent and robust against scalable wavelet-based image coding even at very low bit-rate coding.
Abstract: This paper proposes a novel human vision system based, spread spectrum method to scalable image watermarking. A scalable decomposition of the watermark is spread into the entire frequency sub-bands of the wavelet decomposed image. At each wavelet sub-band, the watermark data are inserted into the selected coefficients of the sub-band in a manner that the watermark embedding visual artifact occurs in the highly textured, highly contrasted and very dark/bright areas of the image. In the lowest frequency sub-band of wavelet transform, the coefficients are selected by independent analysis of texture, contrast and luminance information. In high frequency sub-bands, the coefficient selection is done by analyzing coefficients amplitude and local entropy. The experimental results show that the watermarked test images are highly transparent and robust against scalable wavelet-based image coding even at very low bit-rate coding. The proposed approach can guarantee content authentication for scalable coded images, especially on heterogeneous networks which different users with different process capabilities and network access bandwidth use unique multimedia sources.
5 citations
Cites methods from "SVM regression based robust image w..."
...Some of these methods have been designed as uncompressed domain methods [5, 8, 25, 34, 46, 48, 64] and some others are designed as joint compression-watermarking methods [11, 14, 59]....
[...]
TL;DR: A novel beamforming algorithm (named as SVR-CMT algorithm) is presented for controlling the sidelobes and the nullling level and can improve the output signal-to-interference-and-noise ratio (SINR) performance even if the direction-of-arrival (DOA) errors exist.
Abstract: Minimum variance distortionless response (MVDR) beamformer is an adaptive beamforming technique that provides a method for separating the desired signal from interfering signals. Unfortunately, the MVDR beamformer may have unacceptably low nulling level and high sidelobes, which may lead to significant performance degradation in the case of unexpected interfering signals such as the rapidly moving jammer environments. Via support vector machine regression (SVR), a novel beamforming algorithm (named as SVR-CMT algorithm) is presented for controlling the sidelobes and the nullling level. In the proposed method, firstly, the covariance matrix is tapered based on Mailloux covariance matrix taper (CMT) procedure to broaden the width of nulls for interference signals. Secondly, the equality constraints are modified into inequality constraints to control the sidelobe level. By the ε-insensitive loss function for the sidelobe controller, the modified beamforming optimization problem is formulated as a standard SVR problem so that the weight vector can be obtained effectively. Compared with the previous works, the proposed SVR-CMT method provides better beamforming performance. For instance, (1) it can effectively control the sidelobe and nullling level, (2) it can improve the output signal-to-interference-and-noise ratio (SINR) performance even if the direction-of-arrival (DOA) errors exist. Simulation results demonstrate the efficiency of the presented approach.
3 citations
Cites background from "SVM regression based robust image w..."
...The theory of SVR is initially introduced by Vapnik [17] based on the principle of structural risk minimization and is applied in a number of communications problems such as wireless sensor networks [18] and digital image processing [19]....
[...]
TL;DR: The prime objective of this manuscript is to reveal the research problem and the efficient requirement to implement robust watermarking technique after analyzing the progress of water marking schemes and current research trend.
Abstract: Nowadays digital media has reached the general level of resource sharing system and become a convenient way for sharing lots of information among various individuals. However, these digital data are stored and shared over an internet which is an entirely unsecured and most frequently attacked by several attackers, resulting in a massive loss at various parameters and creates severe issues of copyright protection, ownership protection, authentication, secure communication, etc. In recent years, digital watermarking technology has received extensive attention from users and researchers for content protection and digital data authentication. However, before implementing digital watermarking techniques in practical applications, there are still many problems that need to be solved technically and efficiently. The purpose of this manuscript is to provide a detailed survey on current research techniques of digital watermarking techniques for all media formats with their applications and operational process. The prime objective of this manuscript is to reveal the research problem and the efficient requirement to implement robust watermarking technique after analyzing the progress of watermarking schemes and current research trend.
2 citations
Cites methods from "SVM regression based robust image w..."
...The system then uses the syntactic tool list, WordNet and dictionary and generates the proper watermarked sentence with the help of secret-key [24]....
[...]
References
More filters
18 Aug 2009
TL;DR: Experiments show that the proposed RST invariant watermarking scheme can effectively resist JPEG compression, noise and geometric attacks.
Abstract: A rotation, scaling and translation (RST) invariant watermarking scheme using support vector machine and image moments for synchronization is proposed in this paper. Watermark is adaptively embedded in discrete wavelet transform (DWT) domain. In order to estimate RST transform parameters, support vector machine is utilized to learn image geometric pattern represented by six combined low order image moments. Watermark extraction is carried out after watermarked image has been synchronized. Experiments show that the proposed scheme can effectively resist JPEG compression, noise and geometric attacks.
15 citations
"SVM regression based robust image w..." refers methods in this paper
...Whereas in scheme [14], six combined low order moments were used to train the SVM model for estimating rotation , scaling and translation attacks....
[...]
Posted Content•
TL;DR: An improved steganography approach for hiding text messages in lossless RGB images is proposed and a compression based scheme is introduced to improve the security as well as the image embedding capacity indirectly.
Abstract: This paper proposes an improved steganography approach for hiding text messages in lossless RGB images. The objective of this work is to increase the security level and to improve the storage capacity with compression techniques. The security level is increased by randomly distributing the text message over the entire image instead of clustering within specific image portions. Storage capacity is increased by utilizing all the color channels for storing information and providing the source text message compression. The degradation of the images can be minimized by changing only one least significant bit per color channel for hiding the message, incurring a very little change in the original image. Using steganography alone with simple LSB has a potential problem that the secret message is easily detectable from the histogram analysis method. To improve the security as well as the image embedding capacity indirectly, a compression based scheme is introduced. Various tests have been done to check the storage capacity and message distribution. These testes show the superiority of the proposed approach with respect to other existing approaches.
13 citations
"SVM regression based robust image w..." refers methods in this paper
...The technique of digital watermarking[1], stenography [2] comes in the picture for fulfilling the demand of copyright protection, which embeds the information about ownership claim and authentication into digital data to prevent from illegal and unauthorized copying....
[...]
TL;DR: Experimental results show that the proposed image watermarking is not only invisible and robust against common image processing operations such as filtering, noise adding, and JPEG compression etc., but also robust against the geometrical distortions.
Abstract: Geometric distortion is known as one of the most difficult attacks to resist, for it can desynchronize the location of the watermark and hence causes incorrect watermark detection. It is a challenging work to design a robust image watermarking scheme against geometric distortions. Based on the Support Vector Machine (SVM) geometric distortions correction, we propose a new image watermarking algorithm with good visual quality and reasonable resistance toward geometric distortions in this paper. Firstly, the significant bitplane image is extracted from original host, and DWT is performed on the significant bitplane image. Then, the corresponding low-pass subband is divided into small blocks. Finally, the digital watermark is embedded into host image by adaptively modulating the selected wavelet coefficients in small blocks. The main steps of digital watermark detecting procedure include: (1) the significant bitplane image is extracted from test image, and some low-order pseudo-Zernike moments of the significant bitplane image are computed, which are regarded as the effective feature vectors; (2) the appropriate kernel function is selected for training, and a SVM training model can be obtained; (3) the test image is corrected with the well trained SVM model; (4) the digital watermark is extracted from the corrected test image. Experimental results show that the proposed image watermarking is not only invisible and robust against common image processing operations such as filtering, noise adding, and JPEG compression etc., but also robust against the geometrical distortions.
9 citations
"SVM regression based robust image w..." refers methods in this paper
...The process of geometric distortion correction technique proposed by Wang et al [16] have been utilized in our work....
[...]
...To counter geometric attack, Wang et al [15] propend a technique which uses a the steady Krawtchouk moments as a feature vector....
[...]