QR code (quick response code) is used due to its beneficial properties, especially in the mobile payment field. However, there exists an inevitable risk transaction process. It not easily perceived that attacker tampers with or replaces contains merchant’s beneficiary account. Thus, it of great urgency conduct authentication code. In this study, we propose a novel mechanism based on visual cryptography scheme (VCS) and aesthetic code, which three primary schemes for different concealment...

The widespread availability of digital multimedia data has led to a new challenge in forensics. Traditional source camera identification algorithms usually rely on various traces the capturing process. However, these have become increasingly difficult extract due wide image processing algorithms. Convolutional Neural Networks (CNN)-based demonstrated good discriminative capabilities for different brands and even models devices. their performances is not ideal case distinguishing between...

Quantum superoperator learning is a pivotal task in quantum information science, enabling accurate reconstruction of unknown operations from measurement data. We propose robust approach based on the matrix sensing techniques for that extends beyond positive semidefinite case, encompassing both channels and Lindbladians. first introduce randomized design using near-optimal number measurements. By leveraging restricted isometry property (RIP), we provide theoretical guarantees identifiability...

The majority of deep learning methods for detecting image forgery fail to accurately detect and localize the tampering operations. Furthermore, they only support a single type. Our method introduces three key innovations: (1) A spatial perception module that combines rich model (SRM) with constrained convolution, enabling focused detection traces while suppressing interference from content; (2) hierarchical feature architecture integrates Swin Transformer UperNet effective multi-scale...

Mobile devices have been increasingly used to take pictures without leaving a trace. However, the application system can lead confidential information leaks. A framework for screen-shooting-resilient watermarking via deep networks (SSDeN) in frequency domain is put forward this study solve problem. The proposed extract watermark from leaked photo copyright protection. SSDeN an end-to-end process that combines convolutional neural network (CNN) with residual block embed and watermarks DCT...

With high performances of image capturing tools, information can be easily obtained by screenshots that make copyright protection a challenging task. The existing screen-shooting watermarking algorithms suffer from huge running time, in addition to their low robustness against different screenshot attacks, such as distances and angles the screenshots. In this paper, fast robust high-capacity flexible algorithm for images is proposed. Firstly, Oriented FAST Rotated BRIEF (ORB) feature points...

To protect the copyright of color image, a image watermarking scheme based on quaternion discrete Fourier transform (QDFT) and tensor decomposition (TD) is presented. Specifically, cover partitioned into non-overlapping blocks, then QDFT performed each block. Then, three imaginary frequency components are used to construct third-order tensor. The decomposed by Tucker generates core Finally, an improved odd–even quantization technique employed embed watermark in Moreover, pseudo-Zernike...

Video watermarking plays an important role intoday's copyright protection of video media, in this paper a blind algorithm for mpeg2 is proposed. In algorithm, several DCT intermediate frequency coefficients the pictures are selected and divided into some groups, watermark was embedded by adjusting each group. The capacity adequate experimental result shows proposed robust to compression attack within certain range has little effect on quality video.

Deep learning has become an essential technique in image steganography. Most of the current deep-learning-based steganographic methods process digital images spatial domain. There are problems such as limited embedding capacity and unsatisfactory visual quality. To improve capacity-distortion performance, we develop a method from frequency-domain perspective. We propose module called adaptive channel attention network (AFcaNet), which makes full use frequency features each by fine-grained...

Most of the existing image steganographic approaches embed secret information imperceptibly into a cover by slightly modifying its content. However, modification traces will cause some distortion in stego-image, especially when embedding color data that usually contain thousands bits, which makes successful steganalysis possible. A coverless approach without any for transmitting is proposed. We propose diversity style transfer network using multilevel noise encoding. The consists generator...

In this paper, a new mechanism for digital watermarking is proposed. Three processes are involved: First, the watermarks encapsulated into carrier image; second, sparsity analysis process conducted on one component; finally, flexible selection eventually executed to embed watermarks. Experimental results demonstrate high-capacity information and strong robustness.

Infrared images have been widely used in many research areas, such as target detection and scene monitoring. Therefore, the copyright protection of infrared is very important. In order to accomplish goal image-copyright protection, a large number image-steganography algorithms studied last two decades. Most existing hide information based on prediction error pixels. Consequently, reducing pixels important for steganography algorithms. this paper, we propose novel framework SSCNNP:...

Invoice printing just has two-color printing, so invoice font image can be seen as binary image. To embed watermarks into image, the pixels need to flipped. The more huge watermark is, We proposed a new flipping method in for watermarking capacity. includes one novel interpolation flippable evaluation mechanism, and denoising based on gravity center chaos degree. ensures that keeps features well after scaling. mechanism keep better connectivity smoothness pattern highest structural...

Recently, more and research has focused on the beautification technology of QR (Quick Response) codes. In this paper, a novel algorithm based XOR (exclusive OR) mechanism hybrid basis vector matrices background image synthetic strategy is proposed. The include reverse matrix (RBVM) positive (PBVM). Firstly, RBVM PBVM are obtained by Gauss–Jordan elimination method, according to characteristics RS code. Secondly, modification parity area code can be applied with operation RBVM, used change...

Abstract Autism spectrum disorder (ASD) is a widely recognized neurodevelopmental disorder, yet the identification of reliable imaging biomarkers for its early diagnosis remains challenge. Considering specific manifestations ASD in eyes and interconnectivity between brain eyes, this study investigates through lens retinal analysis. We specifically examined differences macular region retina using optical coherence tomography (OCT)/optical angiography (OCTA) images children diagnosed with...