Invisible Watermarking Research Articles

A Transformer-based invertible neural network for robust image watermarking

For the existing encoder-noise-decoder (END) based watermarking models, since the coupling between the encoder and the decoder is weak, the encoder generally embeds certain redundant features into the cover image to enable the decoder to extract watermark completely, which will affect watermarking invisibility. To address this problem, this paper proposes a Transformer-based invertible neural network (INN) for robust image watermarking (IWFormer). In order to effectively reduce redundant features, the INN framework is utilized for the watermark embedding and extracting processes, so that the encoded features are highly consistent with the features required for decoding. For enhancing watermarking robustness, an affine Transformer module is designed by mining the global correlation of the cover image. In addition, considering that the human visual system is sensitive to low-frequency variations, the wavelet low-frequency sub-band loss is deployed to guide watermark to be embedded in middle- and high-frequency components, thus further increasing the quality of the encoded images. Experimental results demonstrate that compared with the existing state-of-the-art watermarking models, the proposed IWFormer owns remarkable advantages in terms of both watermarking invisibility and robustness.

An effective steganographic technique for hiding the image data using the LSB technique

Steganography is the art and science of writing secret messages so that neither the sender nor the intended recipient knows there is a hidden message. Data hiding is the art of hiding data for various reasons, such as keeping private data, secure, confidential data, etc. With increasing data exchange over a computer network, information security has become a significant issue. There are many methods used for data hiding, and steganography is a well-known technique. Steganography is the art of invisible contact and science. Steganography is the process through which the presence of a message can be kept secret. The objective of this paper is to hide data using the LSB (Least Significant Bit) technique into images that can be detected only by the specified user. We have developed a user-friendly GUI such that it can be used with the utmost ease. This paper is motivated to hide the message stated by the user in the dialog box given within the picture. The secret text is converted to the ciphertext to make it more stable. The sender selects the cover image, and it is used to generate the secured Stegno image, which is identical to the cover image. With the support of a private or public communication network, on the other hand, the stegno image can be saved and sent to the designated user, i.e., the recipient downloads the stegno image and can retrieve the secret text concealed in the stegno image using that same application. As for the watermarking, we have visible and invisible we have used the same LSB technique. In visible watermarking, text or image is embedded in the cover image, which can be noticed easily. As for invisible watermarking, some specific text is inserted into an image, and while retrieving it, it generates a QR code, which can be scanned to get the watermarked text. We used the three different types of cover images i.e. Gray, and RGB also estimated the performance metrics. SNR, MSE, and PSNR the three performance metrics are used, and found that PSNR achieved good results i.e.,71.4733. The RGB image with the hidden text is achieved up to 77.6697

Enhancing quality measurement for visible and invisible watermarking based on M-SVD and DCT

This study introduces an advanced method for evaluating non-blind watermarking quality, leveraging both visible and invisible watermarking techniques grounded in principles of discrete cosine transform (DCT) and modified singular value decomposition (M-SVD). The primary focus is to refine the assessment process of watermarked images by integrating M-SVD, known for its efficacy in measuring image quality and watermarking performance. Results from the M-SVD implementation exhibit a striking resemblance to the original images. The mean squared error (MSE) values for watermarked images range from 0.0003 to 0.0168, while peak signal-to-noise ratio (PSNR) values vary between 42.52 dB and 82.72 dB. These outcomes underscore the potential of DCT and M-SVD techniques in bolstering watermarking processes, especially in invisible watermarking contexts.

Quality Improvement for Invisible Watermarking using Singular Value Decomposition and Discrete Cosine Transform

Image watermarking is a sophisticated method often used to assert ownership and ensure the integrity of digital images. This research aimed to propose and evaluate an advanced watermarking technique that utilizes a combination of singular value decomposition methodology and discrete cosine transformation to embed the Dian Nuswantoro University symbol as proof of ownership into digital images. Specific goals included optimizing the embedding process to ensure high fidelity of the embedded watermark and evaluating the fuzziness of the watermark to maintain the visual quality of the watermarked image. The methods used in this research were singular value decomposition and discrete cosine transformation, which are implemented because of their complementary strengths. Singular value decomposition offers robustness and stability, while discrete cosine transformation provides efficient frequency domain transformation, thereby increasing the overall effectiveness of the watermarking process. The results of this study showed the efficacy of the Lena image technique in gray scale having a mean square error of 0.0001, a high peak signal-to-noise ratio of 89.13 decibels (dB), a universal quality index of 0.9945, and a similarity index structural of 0.999. These findings confirmed that the proposed approach maintains image quality while providing watermarking resistance. In conclusion, this research contributed a new watermarking technique designed to verify institutional ownership in digital images, specifically benefiting Dian Nuswantoro University. It showed significant potential for wider application in digital rights management.

Analysis of healthcare data security with DWT-HD-SVD based-algorithm invisible watermarking against multi-size watermarks

In the modern day, multimedia and digital resources play a crucial role in demystifying complex topics and improving communication. Additionally, images, videos, and documents speed data administration, fostering both individual and organizational efficiency. Healthcare providers use tools like X-rays, MRIs, and CT scans to improve diagnostic and therapeutic capacities, highlighting the importance of these tools in contemporary communication, data processing, and healthcare. Protecting medical data becomes essential for maintaining patient confidentiality and service dependability in a time when digital assets are crucial to the healthcare industry. In order to overcome this issue, this study analyses the DWT-HD-SVD algorithm-based invisible watermarking in medical data. The main goal is to verify medical data by looking at a DWT-based hybrid technique used on X-ray images with various watermark sizes (256*256, 128*128, 64*64). The algorithm’s imperceptibility and robustness are examined using metrics like Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) and are analyzed using Normalized Connection (NC), Bit Error Rate (BER), and Bit Error Rate (BCR) in order to evaluate its resistance to various attacks. The results show that the method works better with smaller watermark sizes than it does with larger ones.

Proactive image manipulation detection via deep semi-fragile watermark

Malicious image tampering refers to intentionally manipulating images to make them harmful to the owners or users. It has become one of the most severe challenges to image authenticity. Conventional methods for detecting tampering by identifying visual artifacts and distortions have limitations due to the rapid advancement of image manipulation techniques, which leave fewer detectable traces. To address these challenges, we propose a proactive media authentication method using deep learning-based semi-fragile watermarks. The designed scheme utilizes deep neural networks to embed an invisible watermark into a target image that is pixel-by-pixel entangled with it, which acts as an indicator of tampering trails. Once the watermarked image is counterfeited, the embedded watermark will exhibit changes accordingly, so we can locate the tampered regions by comparing retrieved and original watermarks. This proactive authentication mechanism makes our method effective against various image tamper techniques, including image copy&move, splicing and in-painting. Although our watermark is designed to be fragile to malicious tampering operations, it remains robust to benign image-processing operations such as JPEG compression, scaling, saturation, contrast adjustments, etc. This design enables our watermark to retain effectiveness when shared over the internet. Extensive experiments demonstrate that our method achieves state-of-the-art forgery detection with superior robustness, imperceptibility and security performance.

Invisible Adversarial Watermarking: A Novel Security Mechanism for Enhancing Copyright Protection

Invisible watermarking can be used as an important tool for copyright certification in the Metaverse. However, with the advent of deep learning, Deep Neural Networks (DNNs) have posed new threats to this technique. For example, artificially trained DNNs can perform unauthorized content analysis and achieve illegal access to protected images. Furthermore, some specially crafted DNNs may even erase invisible watermarks embedded within the protected images, which eventually leads to the collapse of this protection and certification mechanism. To address these issues, inspired by the adversarial attack, we introduce Invisible Adversarial Watermarking (IAW), a novel security mechanism to enhance the copyright protection efficacy of watermarks. Specifically, we design an Adversarial Watermarking Fusion Model (AWFM) to efficiently generate Invisible Adversarial Watermark Images (IAWIs). By modeling the embedding of watermarks and adversarial perturbations as a unified task, the generated IAWIs can effectively defend against unauthorized identification, access, and erase via DNNs, and identify the ownership by extracting the embedded watermark. Experimental results show that the proposed IAW presents superior extraction accuracy, attack ability, and robustness on different DNNs, and the protected images maintain good visual quality, which ensures its effectiveness as an image protection mechanism.

DoBMark: A double-branch network for screen-shooting resilient image watermarking

The dramatic changes in cross-media information transmission modes, especially screen-shooting, have made traditional robust image watermarking for digital channels less resistant to various physical noises from the real world. To address this problem, we propose a double-branch network for screen-shooting resilient image watermarking. Specifically, two weights conforming to a Gaussian distribution are assigned to the double-branch encoder, and by jointly training, two residual images containing watermark information are produced to generate the fused watermarked image. A dual frame alternate-fusion mechanism and a just noticeable difference (JND) constraint function on residual images are employed to improve the watermark invisibility and the quality of watermarked images. A differentiable distortion network is introduced to enhance the robustness of the end-to-end network. Additionally, the decoder integrates a quality enhancement module (QEM) that can correct distortions and apply de-noising to the distorted images, thereby improving the accuracy of watermark information extraction. Experimental results show that the proposed scheme can achieve high robustness against the screen-shooting process while maintaining a satisfactory watermark capacity and visual quality of the watermarked image.

Robust image watermarking using ant colony optimization and fast generic radial harmonic Fourier moment calculation

AbstractIn open networks, the geometric deformation and common image processing are common image manipulation modes, which pose a great challenge in robust watermarking. To improve the robustness of GRHFM‐based watermarking, a watermarking algorithm based on the fast GRHFM calculation method and ant colony optimization (ACO) based fractional parameter selection method is proposed. With the similarity between the discrete Fourier transform and GRHFM moment calculation, the algorithm utilizes fast Fourier transform to improve the GRHFM calculation accuracy and speed. To select the optimal GRHFM fractional parameter for watermarking algorithm, ACO is introduced to the fractional parameter adaptive selection method. Here, the fast Fourier transform‐based calculation method is combined with the adaptive parameter selection method to maximize the invisibility and robustness of watermarking. The experimental results indicate that the algorithm achieves higher robustness with the same payload and invisibility compared with other existing watermarking methods.

High-Quality Evaluation for Invisible Watermarking Based on Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD)

. In this research, we propose an innovative approach that integrates Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) to enhance the quality and security of digital images. The purpose of this technique is to embed imperceptible watermarks into images, preserving their integrity and authenticity. The integration of DCT allows for an efficient transformation of image data into frequency components, forming the basis for embedding watermarks that are nearly invisible to the human eye. In this context, SVD offers an advantage by separating singular values and corresponding vectors, facilitating a more sophisticated watermarking process. The quality evaluation using metrics such as MSE, PSNR, UQI, and MSSIM demonstrates the effectiveness of this approach. Low average MSE values, ranging from 0.0058 to 0.0064, indicate minimal distortion in the watermarked images. Additionally, high PSNR values, ranging from 67.20 dB to 67.22 dB, affirm the high image quality achieved after watermarking. These results validate that the integration of DCT and SVD provides a high level of security while maintaining optimal visual quality in digital images. This approach is highly relevant and effective in addressing the challenges of image protection in this digital era.

Robust Watermarking Algorithm for Building Information Modeling Based on Element Perturbation and Invisible Characters

With the increasing ease of building information modeling data usage, digital watermarking technology has become increasingly crucial for BIM data copyright protection. In response to the problem that existing robust watermarking methods mainly focus on BIM exchange formats and cannot adapt to BIM data, a novel watermarking algorithm specifically designed for BIM data, which combines element perturbation and invisible character embedding, is proposed. The proposed algorithm first calculates the centroid of the enclosing box to locate the elements, and establishes a synchronous relationship between the element coordinates and the watermarked bits using a mapping mechanism, by which the watermarking robustness is effectively enhanced. Taking into consideration both data availability and the need for watermark invisibility, the algorithm classifies the BIM elements based on their mobility, and perturbs the movable elements while embedding invisible characters within the attributes of the immovable elements. Then, the watermark information after dislocation is embedded into the data. We use building model and structural model BIM data to carry out the experiments, and the results demonstrate that the signal-to-noise ratio and peak signal-to-noise ratio before and after watermark embedding are both greater than 100 dB. In addition, the increased information redundancy accounts for less than 0.15% of the original data., which means watermark embedding has very little impact on the original data. Additionally, the NC coefficient of watermark extraction is higher than 0.85 when facing attacks such as translation, element addition, element deletion, and geometry–property separation. These findings indicate a high level of imperceptibility and robustness offered by the algorithm. In conclusion, the robust watermarking algorithm for BIM data fulfills the practical requirements and provides a feasible solution for protecting the copyright of BIM data.

Retracted: Enhanced Invisibility and Robustness of Digital Image Watermarking Based on DWT-SVD

Retracted: Enhanced Invisibility and Robustness of Digital Image Watermarking Based on DWT-SVD

RDD-net: Robust duplicated-diffusion watermarking based on deep network

This paper proposes a novel robust duplicated-diffusion watermarking model based on deep network (RDD-net). RDD-net employs the encoder-noise-decoder structure for end-to-end training to obtain high image watermarking invisibility and robustness. Firstly, a duplicated-diffusion strategy is employed in the encoder to replicate the original watermark iteratively until all copies are diffused to the whole image so that the generalized ability in resisting various noises is obtained. Then, a channel connection technique is designed to extract inherent image features for fusing watermark for robustness by mining correlations of RGB three channels of the color image. Meanwhile, each channel is fused with watermark to increase robustness. Another attempt to improve the watermarking performance is the optimizer, which optimizes the watermark distribution by evaluating the similarities between the encoded image and the original image, as well as between the encoded image and the noised image. Our extensive experimental results demonstrate that the proposed RDD-net not only resists different noises, but also obtains better image quality and higher robustness than the existing watermarking models.

Machine Learning-Assisted Watermarking via Bias Control of Electro-Optical Modulator

We propose an optical watermarking strategy based on precise bias tuning of an electro-optical modulator and a convolutional neural network-enabled decoder. This scheme can be used for identity authentication at the physical layer of optical networks. This method is demonstrated with a transmission demo of about 250 kbit/s watermarking data in a 10-Gbit/s public 20-km communication. The watermarking data is modulated on the bias states of the modulator based on the precise and stable bias control technique. The watermark embedding has been verified to make almost no impact on the 20-km public transmission. For the legal receiver, the invisible watermark data can be extracted with the aid of the convolutional neural network-enabled decoder to verify the legality of the received data. Only one optical source is needed in the proposed scheme and the watermarking is integrated into the public signal naturally. We analyzed and verified the influence of watermarking data on the bit error rate and the channel capacity of the public data. The privacy and security of watermarking data are also analyzed.

Efficient large invisible color watermark embedding using conditional deep autoencoder model for medical applications

The widespread availability of the internet, in conjunction with very low-cost digital recording and storage devices, has ushered in an age in which the reproduction, illegal use, and malicious dissemination of digital information have become much more straightforward. Authentication of multimedia materials has garnered a lot of interest in recent days as a means of preventing illegal usage, theft, and misrepresentation of the content. Invisible watermarking tries to hide information in a medium in order to demonstrate ownership, integrity, or to hide a secret message. The goal of invisible watermarking is to hide the watermark and extract it without making it evident that the cover image is watermarked. This paper presents an invisible watermarking technique that can hide large color water marks in the cover media. The proposed model uses a Conditional Variational Autoencoder (CVAE) to encode the watermark into the cover image. The watermarked image is then decoded at the receiver to extract the watermark. Unlike the conventional watermarking techniques which hide a simple, small black and white image as watermark, the proposed model can embed large full color images as watermarks into the cover images. This makes the proposed model superior to the existing models by hiding large watermarks into the color images. The stego images produced a high PSNR greater than 40 for different watermarks and the images are visually indistinct from the original cover images.

Digital image watermarking using discrete cosine transformation based linear modulation

The proportion of multimedia traffic in data networks has grown substantially as a result of advancements in IT. As a result, it's become necessary to address the following challenges in protecting multimedia data: prevention of unauthorized disclosure of sensitive data, in addition to tracking down the leak's origin, making sure no alterations may be made without permission, and safeguarding intellectual property for digital assets. watermarking is a technique developed to combat this issue, which transfer secure data over the network. The main goal of invisible watermarking is a hidden exchange of data and a message from being discovered by a third party. The objective of this work is to develop a digital image watermarking using discrete cosine transformation based linear modulation. This paper proposed an invisible watermarking method for embedding information into the transformation domain for the grey scale images. This method used the embedding of a stego-text into the least significant bit (LSB) of the Discrete Cosine Transformation (DCT) coefficient by using a linear modulation algorithm. Also, a stego-text is embedded with different sizes ten times within images after embedding the stego-image immune to different kinds of attack, such as salt and pepper, rotation, cropping, and JPEG compression with different criteria. The proposed method is tested using four benchmark images. Also, to evaluate the embedding effect, PSNR, NC and BER are calculated. The outcomes show that the proposed approach is practical and robust, where the obtained results are promising and do not raise any suspicion. In addition, it has a large capacity, and its results are imperceptible, especially when 1bit/block is embedded.

Watermarking in Medical Image

Medical image security is possible using digital watermarking techniques. Important information is included in a host medical image in order to provide integrity, consistency, and authentication in the healthcare information system. This paper introduces a proposed method for embedding invisible watermarking in the 3D medical image. The cover medical image used is DICOM which consists of a number of slices, each one representing a sense, firstly must separate the ROI (Region of Interest) and NROI (Not Region Of Interest) for each slice, the separation process performed by the particular person who selected by hand the ROI. The embedding process is based on a key generated from Arnold's chaotic map used as the position of a pixel in the slices with the highest saturation for embedding a secret message in the NROI because the ROI contains the information of the ill, so it cannot be modified. The evaluation of the proposed method using PSNR and MSE shows good results according to both requirements of the watermark system, the invisibility and quality of the watermarked medical image where the PSNR value is up to 43.3936 and MSE is up to 0.000041.

BHI: Embedded invisible watermark as adversarial example based on Basin-Hopping improvement

Traditional research on generating adversarial examples has mainly focused on artificially adding imperceptible perturbations to input examples. Typically, the perturbation information in adversarial examples has no practical meaning. This study proposes a novel scheme for generating adversarial examples by embedding invisible watermarks based on basin-hopping improvement (BHI). To produce adversarial examples, the proposed BHI scheme is implemented by embedding invisible watermarks into original images based on a data-hiding technique. Specifically, the BHI scheme determines the corresponding coordinates of host images and watermark sizes. By considering the specific coordinates and size, the BHI scheme invisibly embeds the watermarks into the host images to generate adversarial examples. Experimental results show that the attack success rate of the BHI scheme reaches 93.5%. The proposed BHI scheme not only completes the function of an adversarial attack but also protects the copyright of adversarial examples. Moreover, the adversarial examples exhibit outstanding visual and robust performance, providing additional visual protection to avoid the risk of attack.

Hybrid and Invisible Digital Image Watermarking Technique Using IWT-DCT and Hopfield Neural Network

According to the characteristic of HVS (Human Visual System) and the association memory ability of the neural network, an adaptive image watermarking algorithm based on a neural network is proposed invisible image watermarking is a secret embedding scheme for hiding secret images into a cover image file and the purpose of invisible watermarking is copyrights protection. Wavelet transformation-based image watermarking techniques provide better robustness for statistical attacks in comparison to Discrete Cosine Transform domain-based image watermarking. The joined method of IWT (Integer Wavelet Transform) and DCT (Discrete Cosine Transform) gives benefits to the two procedures. The IWT has an impediment of portion misfortune in embedding which increments the mean square estimate as SIM and results in diminishing PSNR. The capacity of drawing in is improved by pretreatment and re-treatment of image scrambling and Hopfield neural network. The proposed algorithm presents a hybrid integer wavelet transform and discrete cosine transform-based watermarking technique to obtain increased imperceptibility and robustness compared to the IWT-DCT-based watermarking technique. The proposed watermarking technique reduces the fractional loss compared to DWT-based watermarking.

A bilateral attention based generative adversarial network for DIBR 3D image watermarking

This paper presents a bilateral attention based generative adversarial network (BAGAN) for depth-image-based rendering (DIBR) 3D image watermarking to protect the image copyright. Convolutional block operations are employed to extract main image features for robust watermarking, but embedding watermark into some features will degrade image quality much. To relieve this kind of image distortion, the bilateral attention module (BAM) is utilized by mining correlations of the center view and the depth map to compute attention of the 3D image for guiding watermark to distribute over different image regions. Since a modality gap exists between the center view and the depth map, a cross-modal feature fusion module (CMFFM) is designed for BAM to bridge the cross-view gap. Because the depth map has lots of flat background information including many redundant features, to prune them, the depth redundancy elimination module (DREM) is used for cross-view feature fusion. In the decoder, two extractors with the same structure are built to recover watermark from the center view and the synthesized view, respectively. In addition, the discriminator is supposed to build a competitive relationship with the encoder to increase the image quality. The noise sub-network is used to train different image attacks for robustness. Extensive experimental results have demonstrated that the proposed BAGAN can obtain higher watermarking invisibility and robustness compared with existing DIBR 3D watermarking methods. Ablation experiments have also proven the effectiveness of DREM, CMFFM and BAM on BAGAN.