Search

Image steganography is the art of concealing secret information within images to prevent detection. In deep‐learning‐based image steganography, a common practice is to fuse the secret image with the cover image to directly generate the stego image. However, not all features are equally critical for data hiding, and some insignificant ones may lead to the appearance of residual artifacts in the stego image. In this article, a novel network architecture for image steganography with hybrid attention mechanism based on generative adversarial network is introduced. This model consists of three subnetworks: a generator for generate stego images, an extractor for extracting the secret images, and a discriminator to simulate the detection process, which aids the generator in producing more realistic stego images. A specific hybrid attention mechanism (HAM) module is designed that effectively fuses information across channel and spatial domains, facilitating adaptive feature refinement within deep image representations. The experimental results suggest that the HAM module not only enhances the image quality during both the steganography and extraction processes but also improves the model's undetectability. Stego images are mixed with varying levels of noise in the training process, which can further improve robustness. Finally, it is verified that the model outperforms current steganography approaches on three datasets and exhibits good undetectability. [ABSTRACT FROM AUTHOR]

Image hiding is a task that hides secret images into cover images. The purposes of image hiding are to ensure the secret images are invisible to the human and the secret images can be recovered. The current state‐of‐the‐art steganography methods run the risk of secret information leakage. A safe image hiding network (SIHNet) is presented to reduce the leakage of secret information. Based on some phenomena of image hiding methods which use invertible neural network, a reversible secret image processing (SIP) module is proposed to make the secret images suitable for hiding and make the stego images leak less secret information. Besides, a reversible lost information hiding (LIH) module is used to hide the lost information into the cover images, thus the method can recover the secret images better than the method that uses random noise to replace the lost information. Experimental results show that SIHNet outperforms other state‐of‐the‐art methods on the PSNR and SSIM values of the recovered secret images and the stego images. Besides, residual images of other state‐of‐the‐art methods all contain information about secret images while residual images of SIHNet leak almost no secret information. Thus the method can prevent the listener of transmission channel from obtaining the information of the secret image through the residual image, which means SIHNet performs better in security than other state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]

Elucidating the mechanism of soil erosion caused by extreme rainstorms could improve agroecosystems' resilience to the extreme climate. However, less attention has been paid to the regional difference of soil erosion during extreme rainstorms through multiple field investigations. Using six field surveys of extreme rainstorms, soil erosion and its driving mechanism were investigated on farmland in the Loess area and the Rocky mountain area. Conclusions are as follows: (1) Farmland was severely eroded in comparison with abandoned land and forest–grass land. The erosion modulus of farmland in the Loess area (51 606.51 t/km2) was greater than that in the Rocky mountain area (2582.75 t/km2). (2) In both the Loess area and the Rocky mountain area, the main erosion types on sloped cropland and terraced land were rills/gullies and embankment failure, respectively. Terracing is an appropriate measure that can effectively control farmland erosion in rainstorm events. (3) Concentrated flow is a major driver of soil erosion, with nearly all rills/gullies occurring in the flow paths. Drainage measures can prevent the formation of concentrated flow, weakening rills/gullies erosion in farmland. (4) Building bio‐embankment, planting high‐coverage crops, and implementing local protection measures could reduce the farmland erosion influenced by extremely heavy rainfall. Combining drainage and agronomic measures is the common strategy to combat extreme rainstorm events for reducing farmland erosion in the two regions. These results can be applied to improve the ability of agricultural systems to cope with extreme rainstorms. [ABSTRACT FROM AUTHOR]

Image preprocessing models are usually employed as the preceding operations of high‐level vision tasks to improve the performance. The adversarial attack technology makes both these models face severe challenges. Prior research is focused solely on attacking single object detection models, without considering the impact of the preprocessing models (multifocus image fusion) on adversarial perturbations within the object detection system. Multifocus image fusion models work in conjunction with the object detection models to enhance the quality of the images and improve the capability of object detection system. Herein, the problem of attacking object detection system that utilizes multifocus image fusion as its preprocessing models is addressed. To retain the attack capabilities of adversarial samples against as many perturbations as possible, new attack method called adaptive retention attack (ARA) is proposed. Additionally, adversarial perturbations concentration mechanism and image selection mechanism, which, respectively, enhance the transferability and attack capability of ARA‐generated adversarial samples. Extensive experiments have demonstrated the feasibility of the ARA. The results confirm that the ARA method can successfully bypass multifocus image fusion models to attack the object detection model. [ABSTRACT FROM AUTHOR]

Electrochemical carbon dioxide reduction reaction (CO2RR) using clean electric energy provides a sustainable route to generate highly‐valuable chemicals and fuels, which is beneficial for realizing the carbon‐neutral cycle. Up to now, achieving a narrow product distribution and highly targeted product selectivity over Cu‐based electrocatalysts is still a big challenge. Herein, sulfur modification on different crystal planes of cuprous oxide (Cu2O) is demonstrated, results in an improvement for formate generation to different degrees. Experimental results and density functional theory (DFT) calculations reveal that sulfur species modified on the surface of Cu2O (100) facet effectively lower the formation energy of key intermediate *OCOH for formate generation compared with Cu2O (111) facet. As a consequence, the modification of p‐Block elements over Cu‐based electrocatalysts is an effective strategy to optimize the adsorption energy of key intermediate during CO2RR, leading to a highly selective product. [ABSTRACT FROM AUTHOR]

Based on iterated Crank–Nicolson (CN) procedure, an alternative algorithm with perfectly matched layer (PML) formulation is proposed in the body‐of‐revolution (BOR) finite‐difference time‐domain (FDTD) lattice for the simulation of rotational symmetric geometrics. For the nonuniform domain simulation, an alternative subgridding method is employed to during the simulation. The iterated CN procedure improves the efficiency through preventing the calculation of tri‐diagonal matrices. The alternative subgridding method enhances the accuracy in nonuniform domains by the calculation of subregions. Numerical example is carried out for the demonstration of effectiveness including efficiency, accuracy and absorption. Through the results, the proposed scheme shows considerable absorption and accuracy improvement in nonuniform domains. Compared with the other CN schemes, the iterated CN procedure can significantly increase the efficiency with small time steps. In conclusion, the advantages and novelty of the proposed algorithm can be described as follows: (1) The iterated CN procedure is proposed for rotational symmetric geometrics. (2) Absorption boundary condition for iterated CN is proposed in BOR‐FDTD. (3) An alternative subgridding method for iterated CN procedure is proposed in BOR‐FDTD lattice. Thus, the proposed algorithm shows potential in nonuniform rotational symmetric geometrics open region simulation. [ABSTRACT FROM AUTHOR]

By incorporating higher order formulation and one‐step leapfrog alternating direction implicit (ADI) procedure, unconditionally stable perfectly matched layer implementation is proposed for terminating frequency‐dependent left‐handed material in finite‐difference time‐domain lattice. To be more specific, frequency‐dependent left‐handed material can be calculated by the piecewise linear recursive convolution method. The proposed scheme shows advantages of higher order formulation and one‐step leapfrog ADI procedure in terms of enhancing absorbing performance, improving computational efficiency and simulating open region problems. The effectiveness is demonstrated through the computational domain with dipole antenna model above the left‐handed material slab. From results, it can be concluded that the proposed formulation can obtain considerable absorption, outstanding efficiency, and improved entire performance. Meanwhile, the proposed scheme also keeps its stability when time step far exceeds the Courant‐Friedrich‐Levy condition. [ABSTRACT FROM AUTHOR]

Electrochemically reducing carbon dioxide (CO2RR) to ethylene is one of the most promising strategies to reduce carbon dioxide emissions and simultaneously produce high value‐added chemicals. However, the lack of catalysts with excellent activity and stability limits the large‐scale application of this technology. In this work, a graphitic carbon nitride (g‐C3N4)‐supported Cu2O composite was fabricated, which exhibited a 32.2 % faradaic efficiency of C2H4 with a partial current density of −4.3 mA cm−2 at −1.1 V vs. reversible hydrogen electrode in 0.1 m KHCO3 electrolyte. The introduction of g‐C3N4 support not only enhanced the uniform dispersion of Cu2O nanocubes, but also stabilized the important *CO intermediates. Moreover, the g‐C3N4 itself had a good activity of reducing CO2 to form *CO, which enriched the key intermediates of C−C coupling around cuprous oxide. The findings highlight the importance of the g‐C3N4 support, a unique two‐dimensional material, including not only the strong CO2 adsorption and activation capacity but also its synergistic effect with the cuprous oxide in CO2RR selectivity. [ABSTRACT FROM AUTHOR]

Electrochemical conversion of CO2 into valuable products by utilizing renewable electricity is a thriving research field. For electrochemical reduction of CO2 to CO, uniform sponge‐like porous Ag nanocubes (SPC‐Ag) integrated on carbon paper are prepared by using an electrochemically driven method. The SPC‐Ag has a 3 D porous structure and a large specific surface area, which affords abundant active sites for the CO2 reduction reaction (CO2RR), as well as reducing impedance to accelerate the CO2RR kinetics. This distinctive organization affords SPC‐Ag with outstanding electrocatalytic performance for CO2 reduction to CO. High faradaic efficiency (>90 %) and large partial current density for CO with excellent durability are observed in a wide potential window, with a maximum value of 93 % at −0.9 V versus reversible hydrogen electrode. [ABSTRACT FROM AUTHOR]