Presentation attack detection based on two-stream vision transformers with self-attention fusion

Aiming at the performance degradation of the existing presentation attack detection methods due to the illumination variation, a two-stream vision transformers framework (TSViT) based on transfer learning in two complementary spaces is proposed in this paper. The face images of RGB color space and multi-scale retinex with color restoration (MSRCR) space are fed to TSViT to learn the distinguishing features of presentation attack detection. To effectively fuse features from two sources (RGB color space images and MSRCR images), a feature fusion method based on self-attention is built, which can effectively capture the complementarity of two features. Experiments and analysis on Oulu-NPU, CASIA-MFSD, and Replay-Attack databases show that it outperforms most existing methods in intra-database testing and achieves good generalization performance in cross-database testing.     

Large eddy simulation of flow over square cylinder arrays in an octagonal configuration at subcritical Reynolds numbers

In this study, fluid flow over an array of eight, 0.029 m × 0.029 m, square cross‐section cylinders in an octagonal configuration is studied numerically. The mean force coefficients (drag and lift) and the vortex formation characteristics of the array are calculated numerically by utilizing a three‐dimensional large eddy simulation mathematical model for turbulence. The numerical simulation is performed with commercial software ANSYS Fluent 19R1. To investigate the parametric influences, three spacings between the cylinders (0.07, 0.14, and 0.2 m), two array attack angles (0° and 15°), and two Reynolds numbers (4060 and 45 800) are considered. The results comprise flow patterns and force coefficients' variations with Reynolds numbers. The lift force of the downstream cylinder reaches its maximum at α = 15°, and the drag force of the upstream cylinders finds its peak at α = 0°. It is observed through velocity and viscosity contour plots that vortex formation length near the cylinder increases at higher Reynolds number. Velocity vector plots are also presented to show fluid flow behavior near the cylinder. Furthermore, the predicted mean forces on the cylinders are slightly different for different Reynolds numbers, spacings, and angles of attack.     

一种标准模型下无证书签名方案的安全性分析与改进

无证书签名具有基于身份密码体制和传统公钥密码体制的优点，可解决复杂的公钥证书管理和密钥托管问题.Wu和Jing提出了一种强不可伪造的无证书签名方案，其安全性不依赖于理想的随机预言机.针对该方案的安全性，提出了两类伪造攻击.分析结果表明，该方案无法实现强不可伪造性，并在"malicious-but-passive"的密钥生成中心攻击下也是不安全的.为了提升该方案的安全性，设计了一个改进的无证书签名方案.在标准模型中证明了改进的方案对于适应性选择消息攻击是强不可伪造的，还能抵抗恶意的密钥生成中心攻击.此外，改进的方案具有较低的计算开销和较短的私钥长度，可应用于区块链、车联网、无线体域网等领域.     

GreyFan:一种Wi-Fi隐蔽信道攻击框架

针对现实网络中诸如侧信道攻击、HID 攻击等传统的物理攻击，物理隔离被认为是一种较为彻底的抵御网络攻击的安全防护手段。2018 年，业界首次提出了一种物理隔离环境下的 Wi-Fi 隐蔽信道方法——Ghost Tunnel，即在Wi-Fi尚未连接的状态下，无线AP可成功将数据传给发起连接请求的计算机。提出了一种基于Ghost Tunnel方法的攻击框架—— GreyFan，利用该攻击框架攻击者可以对未连接Wi-Fi的用户实施无感知攻击，如文件隐蔽传输、任意代码执行等，并分析了相应的防御技术。     

基于人体关节点数据的攻击性行为识别

针对人体攻击性行为识别问题，提出一种基于人体关节点数据的攻击性行为识别方法。首先，利用OpenPose获得单帧图像中的人体关节点数据，并通过最近邻帧特征加权法和分段多项式回归完成由人体自遮挡和环境因素所导致缺失值的补全；然后，对每个人体定义动态"安全距离"阈值，如果两人真实距离小于阈值，则构建行为特征矢量，其中包括帧间人体重心位移、人体关节旋转角角速度和发生交互时的最小攻击距离等；最后，提出改进的LightGBM算法w-LightGBM，并对攻击性行为进行识别。采用公共数据集UT-interaction对所提出的攻击性行为分类识别方法进行测试实验，准确率达到95.45%。实验结果表明，所提方法能够有效识别各种角度的攻击性行为。     

Key-dependent side-channel cube attack on CRAFT

CRAFT is a tweakable block cipher introduced in 2019 that aims to provide strong protection against differential fault analysis. In this paper, we show that CRAFT is vulnerable to side-channel cube attacks. We apply side-channel cube attacks to CRAFT with the Hamming weight leakage assumption. We found that the first half of the secret key can be recovered from the Hamming weight leakage after the first round. Next, using the recovered key bits, we continue our attack to recover the second half of the secret key. We show that the set of equations that are solvable varies depending on the value of the key bits. Our result shows that 99.90% of the key space can be fully recovered within a practical time.     

A 360-Degree Panoramic Image Inpainting Network Using a Cube Map

Inpainting has been continuously studied in the field of computer vision. As artificial intelligence technology developed, deep learning technology was introduced in inpainting research, helping to improve performance. Currently, the input target of an inpainting algorithm using deep learning has been studied from a single image to a video. However, deep learning-based inpainting technology for panoramic images has not been actively studied. We propose a 360-degree panoramic image inpainting method using generative adversarial networks (GANs). The proposed network inputs a 360-degree equirectangular format panoramic image converts it into a cube map format, which has relatively little distortion and uses it as a training network. Since the cube map format is used, the correlation of the six sides of the cube map should be considered. Therefore, all faces of the cube map are used as input for the whole discriminative network, and each face of the cube map is used as input for the slice discriminative network to determine the authenticity of the generated image. The proposed network performed qualitatively better than existing single-image inpainting algorithms and baseline algorithms.     

Difference angle quantization index modulation scheme for image watermarking

In this paper, we propose a new angle quantization index modulation (AQIM) method, called the difference AQIM (DAQIM) method. The proposed method aims to improve the watermarking performance against gain attacks. Unlike the original AQIM method (Ourique et al., Angle QIM: a novel watermark embedding scheme robust against amplitude scaling distortions, in: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 2, 2005, pp. 797–800), the DAQIM method quantizes the difference of the two angles instead of the angles themselves. The main advantage of the DAQIM method is to disperse the interference to the watermark signal from one angle to more angles. Thus, the watermark has a higher ability to resist attacks. We perform a theoretical analysis of the document-to-watermark ratio (DWR) based on our proposed method. We show that our proposed method can obtain a lower embedding distortion compared to the AQIM and the gradient direction watermarking (GDWM) (Nezhadarya et al., IEEE Trans. Inf. Forensics Secur., 6(4), 2011, 1200–1213), methods under the same robustness and payload conditions. The experimental results demonstrate that our proposed method outperforms common existing methods in terms of the robustness against various attacks such as the JPEG quantization noise, additive white Gaussian noise (AWGN), cropping effect and mean filtering.     

Ni–Fe nanocubes embedded with Pt nanoparticles for hydrogen and oxygen evolution reactions

Electrochemical water-splitting is widely regarded as one of the essential strategies to produce hydrogen energy, while Metal-organic frameworks (MOFs) materials are used to prepare electrochemical catalysts because of its controllable morphology and low cost. Herein, a series of trimetallic porous Pt-inlaid Ni–Fe nanocubes (NCs) are developed with bifunctions of hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). In the process of prepare the electrochemical catalysts, Pt nanoparticles are uniformly embedded in the Fe–Ni PBA cube structure, and ascorbic acid is employed as a reducing agent to reduce Pt²⁺ to Pt nanoparticles. In this work, the cubic structure of Fe–Ni PBA is maintained and the noble metal Pt nanoparticles are embedded. Remarkably, the formation of PBA cubes, Pt inlay and reduction are completed in one step, and Pt nanoparticles are embedded by a simple method for the first time. By employing acid etching method, a porous structure is formed on the PBA cube, which increases the exposed area of the catalyst and provides more active sites for HER and OER. Due to the porous structure, highly electrochemical active surface area and the embedded of highly dispersed Pt nanoparticles, the porous 0.6 Ni–Fe–Pt nanocubes (NCs) exhibits excellently electrocatalytic performance and durable stability to HER and OER. In this work, for HER and OER, the Tafel slopes are 81 and 65 mV dec⁻¹, the overpotential η at the current density of 10 mA cm⁻² are 463 and 333 mV, and the onset potential are 0.444 and 1.548 V, respectively. And after a 12-h i-t test and 1000 cycles of cyclic voltammetry (CV), it maintained high stability and durability. This work opens up a new preparation method for noble metal embedded MOF materials and provided a new idea for the preparation of carbon nanocomposites based on MOF.     

一种军棋机器博弈的多棋子协同博弈方法

针对在军棋博弈不完全信息对弈中，面对棋子不同价值、不同位置、不同搭配所产生的不同棋力，传统的单子意图搜索算法，不能满足棋子之间的协同性与沟通性，同时也缺乏对敌方的引诱与欺骗等高级对抗能力。本文提出一种结合UCT搜索策略的高价值棋子博弈方法，实现高价值棋子协同博弈的策略。实战经验表明:高价值多棋子军棋协同博弈策略优于单棋子军棋博弈策略。