Watermarking Technique Research Articles

A Black-Box Watermarking Modulation for Object Detection Models

Training a Deep Neural Network (DNN) from scratch comes with a substantial cost in terms of money, energy, data, and hardware. When such models are misused or redistributed without authorisation, the owner faces significant financial and intellectual property (IP) losses. Therefore, there is a pressing need to protect the IP of Machine Learning models to avoid these issues. ML watermarking emerges as a promising solution for model traceability. Watermarking has been well-studied for image classification models, but there is a significant research gap in its application to other tasks like object detection, for which no effective methods have been proposed yet. In this paper, we introduce a novel black-box watermarking method for object detection models. Our contributions include a watermarking technique that maps visual information to text semantics and a comparative study of fine-tuning techniques’ impact on watermark detectability. We present the model’s detection performance and evaluate fine-tuning strategies’ effectiveness in preserving watermark integrity.

A NEW NESTED HYBRID DWT-HD-SVD WATERMARKING SCHEME FOR DIGITAL IMAGES

Digital watermarking encrypts private information with the original data to protect the ownership rights of the digital asset. This paper suggests a brand-new nested watermarking technique (two-level watermarking technique) to protect copyrights of digital images. It based on a hybrid discrete wave transformation-Hessenberg decomposition-Singular value decomposition (DWT-HD-SVD). The gray-watermark image is first split into n parts, and each portion goes through two steps of watermarking, using two host images, using hybrid DWT-HD-SVD method and one final level of encryption (based on the Lightweight algorithm, the high-speed cryptographic method). The results of the simulation reveal that the suggested system offers a high level of imperceptibility, reliability and robustness. Also it can be used for both the colored and the grayed images.

Implementation of Robust and Secure Watermarking Algorithm on FPGA using DCT

Digital watermarking deals with embedding digital content in a cover signal so that it becomes indiscernible to make it robust against different security threats. This work demonstrates the implementation of a robust and secure watermarking technique on a Field Programmable Gate Array (FPGA) using Discrete Cosine Transform (DCT). Block by block DCT of the cover image was computed and watermark pixels in each of the blocks were hidden using middle-frequency band coefficients. Security is ensured by encrypting the watermarkwith a secret key and Arnold transform. The resultant watermark image is robust to various threats like JPEG compression, scaling, cropping, and noise attacks. This watermarking approach requires a large amount of data to be processed at high speed. FPGA is the best choice for such an application. Also, it is reprogrammable and easily upgradable according to user application. In this work, a watermarking algorithm is implemented using Xilinx design tools on the Artix-7 FPGA board.

Blockchain for video watermarking: An enhanced copyright protection approach for video forensics based on perceptual hash function.

As a direct result of advancements in digital technology and the Internet, the copyright protection and information integrity of multimedia that are being published across the Internet have emerged as a major and urgent issue that needs to be addressed. The technique of digital watermarking may be used to protect intellectual property. In terms of authentication, resilience, storage, and capacity of digital watermarking information, there is still room for development. Blockchain's potential in video copyright protection and management applications has motivated researchers. Copyright owners and consumers may now communicate directly via copyright protection apps built on the blockchain, eliminating the need for distributers and the associated fees. Nonetheless, the current blockchain-based video watermarking solutions require storing a significant number of coordinates depending on the watermark size and are susceptible to video frame attacks on the video frame texture region. This study proposes an enhanced video copyright management approach that incorporates digital watermarking, the blockchain, and a perceptual hash function. Watermark information is stored on a blockchain structure, which also acts as a timestamp for verification purposes. To verify watermark data without the source video, a perceptual hash function is employed to compute a hash value based on the structural information of video frames. The contribution is in learning how to extract a hash function from a small number of video frames that still adequately represent a large amount of video content while also reducing the number of unnecessary video frames and the amount of computation required to summarize and index that content in a blockchain. This expedites the dissemination of copyrighted works and increases their security and readability, hence facilitating their circulation on the Internet. Our experimental results demonstrate that this approach is memory efficient, as it only needs to store one key for each key frame, regardless of the size of the watermark. Additionally, the overall robustness is greatly improved by using the blockchain's random hash function. Therefore, new and important advancements in video watermarking have been realized because of this effort.

A Compressed Domain Robust Video Watermarking Resists Geometric Attack

The widespread utilization of mobile devices and the proliferation of video content, facilitated by communication networks, has unfortunately led to issues such as data piracy, illegal usage, and unauthorized sharing. To address these challenges, digital watermarking emerges as a crucial tool, serving purposes such as authentication, maintaining integrity, and identifying rightful owners. In the realm of video watermarking, two primary categories exist: watermarking in the compressed domain and watermarking in the uncompressed domain. Compressed domain watermarking involves expedited data processing through partial video decoding, whereas uncompressed domain watermarking requires more time. This article proposes a novel video watermarking methodology designed specifically to withstand geometric attacks. The proposed method aims to dynamically embed watermarks within video content, thereby enhancing resilience against various geometric distortions. Geometric attacks, including scaling, rotation, cropping, and translation, pose significant challenges in digital watermarking and are thus a focal point of this research. The objective is to integrate this approach with complex video coding standards, leveraging compression domain techniques to fortify resistance against geometric attacks. Tests show that the suggested video watermarking system works well. Not only does it excel in safeguarding against geometric attacks, but it also mitigates perceptual quality degradation, thus ensuring a balance between robustness and visual fidelity. As the prevalence of video content continues to grow, robust watermarking techniques like the one presented here play a vital role in preserving intellectual property rights and ensuring the integrity of multimedia content in the digital age.

Secure and Robust Dual Watermarking for Western Blot Images

The rapid growth of digital data sharing has created a pressing need to safeguard the integrity, authenticity, and security of bioanalytical images, such as Western blots, online. To address this challenge, we propose a novel hybrid biometric watermarking technique that combines randomised singular value decomposition (RSVD), discrete wavelet transform (DWT), schur transform, pseudo magic cube, and chaotic encryption, making the watermarking of Western blot images more secure, robust, and imperceptible. Our approach surpasses state-of-the-art methods, ensuring imperceptible watermarking and robust protection against tampering and unauthorised use. By integrating spatial and transform-domain techniques, we achieve superior performance, outperforming existing methods with a peak signal-to-noise ratio (PSNR) of 48.7581, structural similarity index (SSIM) of 0.9998, a normalised correlation (NC) of 0.9996, an average number of changing pixel rate (NPCR) of 0.9961, and an average unified average changing intensity (UACI) of 0.3334. Our technique sets a new standard for digital image protection in scientific research, addressing critical challenges in tampering, copyright protection, and data security. With its high performance and security, our approach enables the trustworthy sharing and publication of Western blot images in the digital era.

Quantum color image watermarking scheme based on quantum discrete cosine transform

Quantum watermarking is a technique that embeds specific information into a quantum carrier, used for digital copyright protection. This paper introduces a new quantum color image watermarking method that combines Quantum LOG (QLOG) edge detection with Quantum Discrete Cosine Transform (QDCT) technology. First, edge detection is used on the RGB channels of the quantum color image. This helps identify the best channel for embedding the watermark. Next, the image is decomposed using the quantum Haar wavelet transform (QHWT). The image is then processed with QDCT, selecting the mid-frequency parts for watermark embedding. Using QDCT improves the watermark’s resistance to JPEG compression. The invisibility of the watermark is verified by assessing its Peak Signal-to-Noise Ratio (PSNR) and histogram. Comparative analysis shows strong resistance to various noise attacks. Simulation results confirm that the watermarking technique maintains high visual quality and resists different adversarial attacks.

Social Media Authentication and Combating Deepfakes using Semi-fragile Invisible Image Watermarking

With the significant advances in deep generative models for image and video synthesis, Deepfakes and manipulated media have raised severe societal concerns. Conventional machine learning classifiers for deepfake detection often fail to cope with evolving deepfake generation technology and are susceptible to adversarial attacks. Alternatively, invisible image watermarking is being researched as a proactive defense technique that allows media authentication by verifying an invisible secret message embedded in the image pixels. A handful of invisible image watermarking techniques introduced for media authentication have proven vulnerable to basic image processing operations and watermark removal attacks. In response, we have proposed a semi-fragile image watermarking technique that embeds an invisible secret message into real images for media authentication. Our proposed watermarking framework is designed to be fragile to facial manipulations or tampering while being robust to benign image-processing operations and watermark removal attacks. This is facilitated through a unique architecture of our proposed technique consisting of critic and adversarial networks that enforce high image quality and resiliency to watermark removal efforts, respectively, along with the backbone encoder-decoder and the discriminator networks. This allows images shared over the Internet to retain the verifiable watermark as long as facial manipulations or any other Deepfake modification technique is not applied. Thorough experimental investigations on SOTA facial Deepfake datasets demonstrate that our proposed model can embed a \(64\) -bit secret as an imperceptible image watermark that can be recovered with a high-bit recovery accuracy when benign image processing operations are applied while being non-recoverable when unseen Deepfake manipulations are applied. In addition, our proposed watermarking technique demonstrates high resilience to several white-box and black-box watermark removal attacks. Thus, obtaining state-of-the-art performance.

Image sterilization via overwriting

Sterilization refers to the method of removing hidden information from images, effectively eradicating malicious data embedded within digital media. Existing methods are hampered by dual limitations: insufficient resilience to cutting-edge information hidden methods, coupled with the challenge of image restoration in quality. In response to these challenges, this paper proposes a novel method composed of overwriting attack and diffusion optimization stages. Employing the concept of pixel destruction, the overwriting method conducts a more thorough attack on hidden information while preserving sufficient information for image reconstruction and further diffusion optimization. This strategy is not bound to any particular information hiding method, making it effective against a wide range of steganography and watermarking techniques by systematically disrupting every pixel in the image. Diffusion optimization makes full use of the generative capabilities of diffusion models trained on large datasets, enabling higher quality image enhancements for a variety of scene images.Our method stands out in comparison to state-of-the-art studies, as experimental outcomes demonstrate its capability to maintain superior image integrity while efficiently eliminating hidden information across various steganography and watermarking techniques. Notably, when confronted with deep robust watermarking schemes, our approach exhibits a significant improvement over preceding methods. For instance, the removal rate improves significantly from 39.42% to 87.78% with higher quality restoration when compared to SC-Net. Our method delivers superior outcomes in the task of image sterilization, while presents a new challenge to the robustness of image information hiding methods and contributes to network security.

Bio-mimicking DNA fingerprint profiling for HLS watermarking to counter hardware IP piracy

The multifaceted, multivendor-based global design supply chain induces hardware threats of intellectual property (IP) piracy for modern computing and electronic systems. Current hardware watermarking techniques fall short either in terms of watermark strength (size of covert constraints generated) or number of security layers/variables involved in the security constraints generation process. This paper presents a novel approach for high level synthesis (HLS) watermarking by bio-mimicking DNA fingerprint profiling to counter hardware IP piracy. The proposed approach effectively captures the vital DNA fingerprint profiling phases such as DNA sequencing, DNA fragmentation, fragment replication, DNA ligase, etc. and bio-mimics them to generate a digital watermarking framework. The presented approach has been demonstrated on convolutional layer and JPEG compression-decompression (CODEC) algorithms that are widely used in several medical and machine learning applications. The proposed approach has been thoroughly compared with several state-of-the-art approaches. The proposed approach depicts superior security in the probability of coincidence of up to ~ 104 and tamper tolerance of up to ~ 10368 at 0% overhead as compared to the prior approaches.

Quaternion attention-based JND model for macrophotography image watermarking

With the advancement of imaging technology, macrophotography images (MPIs) have become a popular research topic. Unlike natural images, MPIs often feature sharp foregrounds and blurred backgrounds, leading to distinct perceptual characteristics in estimation. As the number of MPIs grows rapidly, concerns over image quality and security increase. Robust watermarking techniques have been introduced to address these challenges. Just Noticeable Difference (JND) has been widely used in quantization-based watermarking frameworks. However, existing JND models handle each image area with a single-level perceptual attention. Visual attention in Quaternion Discrete Wavelet Transform (QDWT), which can reflect the Multi-level perceptual attention feature. Therefore, we propose a new method called Quaternion Attention-based Just Noticeable Difference model for MPIs Watemarking (QAJnd-MW) for watermarking MPIs. This method uses visual attention mechanisms, recognizing that the HVS is more sensitive to attention regions. We generate a masking effect in the JND field. The input image undergoes QDWT to explore multi-scale features. The multi-scale feature maps, with multi-directional luminance and multi-channel color, help create local and global attention maps, which are fused to form the final attention map. Specifically, considering both attention-based masking effects, the quaternion attention-guided JND model is designed for a robust MPI watermarking framework, aiming to further improve MPI security. Extensive experiments on the MP2020 and Blur Detection datasets show that the proposed model significantly improves robustness against JPEG compression attacks, reducing the bit error rate (BER) by up to 12%. Additionally, the model performs well against other attacks, such as those in online social networks, with lower BER than current state-of-the-art techniques.

A Video Dual-Domain Blind Watermarking Algorithm Based on Hadamard Transform

Addressing the compatibility challenges surrounding the robustness and reversibility of existing video watermarking techniques, this study introduces a novel video dual-domain blind watermarking algorithm leveraging the Hadamard transform. Specifically tailored for H.264 video copyright protection, the algorithm initially organizes video frames and identifies key frames for watermark embedding. Prior to embedding, the robust watermark undergoes coding preprocessing to optimize its integration. Subsequently, a 4×4 block is expanded based on the selected embedding position within the frame, followed by the application of the Hadamard transform to the enlarged block. The 1-bit robust watermark information is then embedded via the coefficient pair located in the first row of the Hadamard coefficient matrix corresponding to the expanded block. Additionally, a reversible watermark, designed to mitigate the distortions introduced during robust embedding, is generated and embedded into the remaining coefficients of the coefficient matrix using reversible embedding techniques. During watermark extraction, the dual-domain watermark can be retrieved exclusively through reversible extraction methodologies by analyzing the size relationship of coefficient pairs, eliminating the need for access to the original video data. To bolster the algorithm’s robustness, a majority-subordinate voting system is devised and implemented, effectively enhancing its resilience. Experimental findings demonstrate that, compared to similar approaches, this algorithm not only enhances the reversibility of video restoration but also exhibits superior robustness and meets the requirements for imperceptibility.

Ecg signal watermarking using QR decomposition.

This study introduces a novel watermarking technique for electrocardiogram (ECG) signals. Watermarking embeds critical information within the ECG signal, enabling data origin authentication, ownership verification, and ensuring the integrity of research data in domains like telemedicine, medical databases, insurance, and legal proceedings. Drawing inspiration from image watermarking, the proposed method transforms the ECG signal into a two-dimensional format for QR decomposition. The watermark is then embedded within the first row of the resulting R matrix. Three implementation scenarios are proposed: one in the spatial domain and two in the transform domain utilizing discrete wavelet transform (DWT) for improved watermark imperceptibility. Evaluation on real ECG signals from MIT-BIH Arrhythmia database and comparison to existing methods demonstrate that the proposed method achieves: (1) higher Peak Signal-to-Noise Ratio (PSNR) indicating minimal alterations to the watermarked signal, (2) lower bit error rates (BER) in robustness tests against external modifications such as AWGN noise (additive white Gaussian noise), line noise and down-sampling, and (3) lower computational complexity. These findings emphasize the effectiveness of the proposed QR decomposition-based watermarking method, achieving a balance between robustness and imperceptibility. The proposed approach has the potential to improve the security and authenticity of ECG data in healthcare and legal contexts, while its lower computational complexity enhances its practical applicability.

A Survey: Digital Image Watermarking-Robustness and Imperceptibility

With the Internet expanding, multimedia security has become a significant problem. As digital image manipulation is getting simpler, the certification of authenticity has become more difficult. Digital watermarking outperforms conventional security methods for protecting photos from attacks. The need for reliable image authentication methods in the contemporary digital era has always been a hot topic among academics and researchers. When transmitting sensitive information today through a regular correspondence route, many hacks are possible. Everyone, therefore, requires their digital data to be non-tampered. Many solutions are available like stenography, cryptography, and watermarking. Watermarking gives promising results when it comes to the authenticity of the images. The robustness and imperceptibility of several existing watermarking techniques are demonstrated in this study and critical analysis is performed. Additionally, the author provides a thorough introduction to key ideas like the transform domain and the spatial domain. This study also offers a thorough analysis of contemporary methods for watermarking digital images while showcasing established practices in various domains. Hence, this research study could potentially serve as a viable resource for researchers seeking to develop effective watermarking techniques for practical applications.

A Survey of Text Watermarking in the Era of Large Language Models

Text watermarking algorithms are crucial for protecting the copyright of textual content. Historically, their capabilities and application scenarios were limited. However, recent advancements in large language models (LLMs) have revolutionized these techniques. LLMs not only enhance text watermarking algorithms with their advanced abilities but also create a need for employing these algorithms to protect their own copyrights or prevent potential misuse. This paper conducts a comprehensive survey of the current state of text watermarking technology, covering four main aspects: (1) an overview and comparison of different text watermarking techniques; (2) evaluation methods for text watermarking algorithms, including their detectability, impact on text or LLM quality, robustness under target or untargeted attacks; (3) potential application scenarios for text watermarking technology; (4) current challenges and future directions for text watermarking. This survey aims to provide researchers with a thorough understanding of text watermarking technology in the era of LLM, thereby promoting its further advancement.

Copyright Protection Based on Hybrid Image Watermark (DCT) with Audio Watermark (EMD)

A digital watermark is a type of security that incorporates data into a signal, making it difficult to erase and showing ownership and validity. An important issue brought about by the growing use of the Internet is illicit copyright infringement and authentication. Previous research has proposed several techniques for this problem but has proven them to be either robust, difficult to detect, or fragile and easy to break. Because of this, in this paper, we use two separate kinds of watermarks—one for image and one for audio—to merge these two purposes into one. The image watermark discrete cosine transform (DCT) is renowned for its robustness. It is embedded in an audio watermarking algorithm based on empirical mode decomposition (EMD), ensuring a safe and fragile embedding process. The audio signal was divided into frames, and each one was decomposed by EMD into intrinsic mode functions (IMFs). The DCT watermark bits and the synchronization codes are embedded into the extrema of the first IMF, a high frequency mode sensitive to any small changes in the audio signal. The experimental outcomes suggest that while using two separate kinds of watermarks may seem like a comprehensive solution, it shows the flexibility of EMD and the limits of the DCT watermark technique in different scenarios.

DEEPMARKNET FOR ROBUST IMAGE AND VIDEO WATERMARKING EMBEDDING AND DETECTION

In the digital age, securing multimedia content against unauthorized use is critical. Traditional watermarking techniques often struggle with robustness against various attacks. This study introduces a novel DeepMarkNet approach for robust image and video watermarking. DeepMarkNet leverages deep learning to embed and detect watermarks with high resilience to common distortions. The method employs a Convolutional Neural Network (CNN) for embedding and a dual- stream architecture for detection. Experimental results demonstrate DeepMarkNet effectiveness, achieving a 98.5% detection accuracy and maintaining watermark integrity under compression and noise attacks. This outperforms conventional techniques by 15% in robustness.

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.

A reversible-zero watermarking scheme for medical images

The paper addresses the issue of ensuring the authenticity and copyright of medical images in telemedicine applications, with a specific emphasis on watermarking methods. While several systems only concentrate on identifying tampering in medical images, others also provide the capacity to restore the tampered regions upon detection. While several authentication techniques in medical imaging have successfully achieved their goals, previous research underscores a notable deficiency: the resilience of these schemes against unintentional attacks has not been sufficiently examined or emphasized in previous research. This indicates the need for further development and investigation in improving the robustness of medical image authentication techniques against unintentional attacks. This research proposes a Reversible-Zero Watermarking approach as a solution to address these problems. The new approach merges the advantages of both the reversible and zero watermarking techniques. This system is comprised of two parts. The first part is a zero-watermarking technique that uses VGG19-based feature extraction and watermark information to establish an ownership share. The second part incorporates this ownership share into the image in a reversible manner using a combination of a discrete wavelet transform, an integer wavelet transform, and a difference expansion. Research findings confirm that the suggested watermarking approach for medical images demonstrates substantial enhancements compared to current methodologies. Research findings indicate that NC values are often around 0.9 for different attacks, whereas BER values are close to 0. It demonstrates exceptional qualities in being imperceptible, distinguishable, and robust. Additionally, the system provides a persistent verification feature that functions independently of disputes or third-party storage, making it the preferred choice in the domain of medical image watermarking.

Assessing the Effectiveness of Network Security Tools in Mitigating the Impact of Deepfakes AI on Public Trust in Media

The rising threat of deepfake technology challenges public trust in media, necessitating robust countermeasures. This study proposes the Anti-DFK framework, a comprehensive strategy to mitigate the spread of deepfakes on major social platforms such as Instagram, Facebook, YouTube, and Twitter. The framework integrates deep learning-based detection engines, digital watermarking, and advanced network access controls, including URL filtering, domain reputation filtering, content-type filtering, and Geo-IP blocking. Analyzing historical deepfake data, user engagement metrics, and public sentiment from Kaggle Datasets, the study employed deep learning models—CNNs, LSTMs, and Transformer-based—to evaluate detection capabilities, achieving the highest controlled environment accuracy of 0.97. Digital watermarking techniques were tested for robustness against various attacks, with the DCT method displaying significant resilience. Network access controls were assessed for their effectiveness in curtailing the spread of deepfakes, with content filtering proving the most effective by reducing dissemination by nearly 80%. Findings indicate a critical negative impact of deepfakes on public trust, underscoring the need for the integrated approach offered by the Anti-DFK framework. The study concludes that implementing these sophisticated detection tools, combined with robust digital watermarking and stringent network controls, can significantly enhance the integrity of media content and restore public confidence.