Visual attention model based dual watermarking for simultaneous image copyright protection and authentication

Ameen

2021

AJRCoS

Digital image authentication techniques have recently gained a lot of attention due to their importance to a large number of military and medical applications, banks, and institutions, which require a high level of security. Generally, digital images are transmitted over insecure media, such as the Internet and computer networks of various kinds. The Internet has become one of the basic pillars of life and a solution to many of the problems left by the coronavirus. As a result, images must be protected from attempts to alter their content that might affect important decision-making. An image authentication (IA) system is a solution to this difficult problem. In the previous literature, several methods have been proposed to protect the authenticity of an image. Digital image watermark is a strategy to ensure the reliability, resilience, intellectual property, and validity of multimedia documents. Digital media, such as images, audio, and video, can hide content. Watermarking of a digital image is a mechanism by which the watermark is embedded in multimedia and the image of the watermark is retrieved or identified in a multimedia entity. This paper reviews IA techniques, watermark embedding techniques, tamper detection methods and discusses the performance of the techniques, the pros and cons of each technique, and the proposed methods for improving the performance of watermark techniques.

Section: Extensive Research On Fragile Watermark As Follow;mentioning

confidence: 99%

Image Authentication Based on Watermarking Approach: Review

Ameen

2021

AJRCoS

“…Digital watermarking involves the embedding of concealed information in the form of a watermark into the host signal, to secure copyrights [6,44], tamper detection [7,8], telemedicine [42,43], content authentication [9,10], source authentication [49], etc. Based on domains, digital watermarking is classified into two groups: spatial and transform/frequency domain techniques [45,46].…”

Section: Introductionmentioning

confidence: 99%

Source identification of shared web content in online social media: A watermarking approach

Shaliyar

Mustafa

2023

Preprint

Purpose An online social media (OSM) is a powerful virtual network that facilitates global forwarding/sharing of messages, planning, analysis, and/or voting. Due to the rapid enhancement of accessibility of internet services, one may simply forward/share web content through OSM. These may include renowned OSM platforms Twitter, WhatsApp, Instagram, and Facebook to name a few. Such a practice of sharing web content without validating the authenticity of the source may have major political, social, or economic ramifications for society. The proposed research work aimed to propose a novel watermarking approach to reveal the first user/source of shared web content (image) on OSM. Method To authenticate the source, the combination of 10 digit mobile number, social security number (AADHAAR number in India), GPS coordinates, and specific code of the messenger app are used as a watermark. Prior to integration, the hamming code is utilized to encode the watermark, to make an approach more robust. In the embedding phase, the cover image is initially split into non-overlapping uniform blocks. Afterward, each block is subjected to Slantlet transformation (SLT). Moreover, four copies of the source-centric data are inserted during the watermark insertion process to achieve high reliability. The proposed method has been validated for effectiveness experimentally and compared with other closely related studies. Results The results revealed a higher level of robustness with a significant level of imperceptibility in terms of BER and PSNR respectively under various signal-processing attacks. In addition, the approach is determined to be fast enough for practical usage. Hence, the identification of the source of the shared content has been achieved to a higher degree. Conclusion A comparison with various existing approaches shows the applicability of the proposed methodology in terms of robustness, durability, and time complexity. The scope of the research will be broadened in the near future to advance in watermarking employing host images of varying sizes, attacks involving rotation and translation, and blockchain technology.

“…To perform watermarking, transform-domain methods usually convert the host image into a transformed domain first, and then hide the watermark information by modifying the transformed coefficients. Commonly employed transformations for image watermarking include discrete Fourier transform (DFT) (Fares et al 2020;Hsu and Hu 2020; Urvoy et al 2014), discrete wavelet transform (DWT) (Barni et al 2001;Hu and Hsu 2017;Huynh-The et al 2016), DCT (Hsu and Hu 2017;Hu et al 2022;Moosazadeh and Ekbatanifard 2019;Patra et al 2010;Singh and Bhatnagar 2018), matrix decomposition (MD) (Ali et al 2015;Chang et al 2005;Hsu and Tu 2020;Koley 2021;Nha et al 2022;Su et al 2020), and combinations of these transforms (Hu and Hsu 2015;Kang et al 2020).…”

Section: Introductionmentioning

confidence: 99%

Blind color image watermarking incorporating a residual network for watermark denoising and super-resolution reconstruction

Hsu²

2022

Soft Comput

Watermarking is a technique for hiding secret information in various types of multimedia data to protect intellectual property rights. Currently, the integration of deep learning technology with image watermarking is reshaping the application and promotion of relevant techniques developed so far. This paper presents a novel type of blind color image watermarking method that embeds a downsized color image into a host color image. Watermarking implementation involves partitioning the host image into non-overlapping blocks of 8 × 8 pixels, performing discrete cosine transform (DCT) for each block of every channel, and then manipulating the magnitudes of three designated DCT coefficients subject to a minimization constraint. The experimental results confirmed that the proposed image watermarking method outperformed four other methods in terms of zero-normalized cross-correlation (ZNCC). Moreover, watermark imperceptibility, as reflected by the measured peak signal-to-noise ratio and mean structural similarity metrics, remained at a satisfactory level. In addition to this new style of color image watermarking, we employed a deep residual network to reduce noise and increase the resolution of the retrieved watermarks. Overall, the residual network achieved a satisfactory ZNCC level (>0.88) when the watermark images were super-resolved by a factor of sixteen.