Embedding Strength Research Articles

Enhancing Robustness and Imperceptibility with a Texture-Based Adaptive QIM Approach

This paper proposes a new approach for image data hiding that aims to improve both robustness and imperceptibility. The proposed approach is based on a texture-adaptive quantization index modulation (QIM) method that takes into account the local characteristics of the image texture to optimize the embedding process. The approach is evaluated using various image datasets and compared to existing state-of-the-art techniques. The results show that the proposed approach achieves better performance in terms of robustness against various image processing attacks, with BERs of zero in some attacks, while maintaining a high level of imperceptibility with PSNRs exceeding 46 decibels. It can be deduced that adaptation and normalization of the embedding strength arguments can enhance the QIM methods’ performance. The proposed approach has potential applications in image authentication, copyright protection, and data hiding.

Experimental Investigation of Full Hole Embedment Behavior of Bamboo Scrimber with Dowel-Type Fasteners

A comprehensive understanding of the embedment behavior is of great importance in the design of contemporary bamboo constructions with connections utilizing dowel-type fasteners. The objective of this research was to assess the embedment behavior of bamboo scrimber using full-hole embedment tests. To investigate the effect of the loading angle and bolt diameter, a series of tests were performed using bolts of varying diameters (16 mm, 18 mm, and 20 mm) and loading angles (0° to 90°, with an increment of 15°). The experimental results demonstrated that the loading angle has a considerable influence on the embedment behavior. As the loading angle was increased, the failure mode underwent a change from a brittle failure mode, which was dominated by shear mechanisms, to a ductile failure mode, which was dominated by fiber crushing. The yield and ultimate embedment strengths showed an M-shaped response to changes in the loading angle, with the lowest values being 0°, 45°, and 90°. The bolt diameter was found to have no impact on the failure mode of the specimen. However, an increase in bolt diameter resulted in a reduction in the embedment strength when the specimen was loaded at 90°.

A DNN robust video watermarking method in dual-tree complex wavelet transform domain

Deep learning is increasingly being applied in the field of robust watermarking. However, the existing deep learning-based video watermarking methods only uses spatial domain information as the input and the robustness against attacks such as H.264/AVC compression is still not strong. Therefore, this paper proposes a deep learning-based robust video watermarking method in dual-tree complex wavelet transform (DT-CWT) domain. The video frames are transformed into the DT-CWT domain and the suitable high-pass subbands are selected as candidate embedding positions. Then, the 2D and 3D convolutions are combined to extract both intra-frame spatial features and inter-frame temporal features for finding the stable and imperceptible coefficients for watermark embedding in the candidate positions. The convolutional attention module (CBAM) is used to further adjust the embedding coefficients and strengths. In addition, the attack layer, where a differentiable proxy is specially designed in this paper for the simulation of non-differentiable H.264/AVC compression, is introduced to generate distorted watermarked videos for improving the robustness against different attacks. Experimental results show that our method is superior to both the existing deep learning-based methods and traditional methods in the robustness against both spatial and temporal attacks while preserving high video quality. The source code is available at https://github.com/imagecbj/A-DNN-Robust-Video-Watermarking-Method-in-DT-CWT-Domain.

Experimental and finite element modelling analysis on the embedment performance and failure mechanisms of flattened-bamboo composite with the effect of fiber orientation angle

This paper focused on a newly engineering use flattened-bamboo composite (FBC) and studied the effect of fiber orientation angle (0° - 90°, 15° intervals) on FBC’s embedment performance and failure modes by half hole test. The results indicated that with larger fiber orientation angle, both the embedment strength and initial stiffness gradually decreased (embedment strength: 52 ∼ 26 MPa, initial stiffness: 27062 ∼ 11149 kN/mm), and Hankinson’s formula could effectively predict FBC’s embedment strength with error less than ±7 %. Combing the experimental phenomenon and scanning electron microscope (SEM) results, the failure modes of embedment specimens were classified. For angles of 0° and 15°, the failure mode was a mix of fiber crushing failure and split failure between fibers. For angles from 30° to 60°, the phenomenon of fiber being crushed became less obvious, only small cracks were witnessed around the half-hole area. For angles of 75° and 90°, the bamboo fiber compression failure perpendicular to fiber direction dominated, and two horizonal cracks happened on both sides of the specimen. A damaged constitutive model using continuum damage mechanics was proposed for simulating FBC’s embedding failure modes, and the proposed model was verified by comparing with the experimental phenomenon.

Exploring the relationship between embedment behaviour in cross-laminated timber and its layers: An experimental study

Cross-laminated timber (CLT) has become popular as a construction material and the design of CLT connections is typically based on empirical equations with an average embedment strength of the wood-based product. The experimental study presented herein aims at enhancing the understanding of the relationship between the CLT embedment behaviour and the single layers it is composed of. Non-linear embedment stress–displacement relationships were measured in 244 full-hole and half-hole embedment tests on structural timber boards as well as on CLT produced of parts of the same boards. Three different fasteners, namely a steel dowel with a diameter of 12 mm and the threaded part of screws with a nominal diameter of 6.5 mm and 10 mm, were studied. The design of the experiments allowed to validate a simple mechanical model with parallel springs for the CLT, which showed good agreement with experiments. In addition to a validation of the non-linear spring model, this unique experimental data-set is further exploited in a comparison with empirical equations and for the derivation of a non-linear embedment ratio, between parallel and perpendicular to the grain embedment stresses, over the displacement. Knowledge of the relationship between the non-linear embedment behaviour of the single layers and the CLT element can be further exploited in the engineering design and numerical modelling of CLT connections with dowel-type fasteners.

Load-bearing capacity of slender dowel-type fasteners in Timber-Concrete Composite connections

Load-bearing capacity of slender dowel-type fasteners of round ring-shank nails and U-shaped connectors (U-conn:s) of square cross-sectional nail thread used as connections in Timber-Concrete Composite structures have been studied, experimentally and with two different theoretical models. Tensile strength and withdrawal capacity of fasteners and embedment strength of timber were tested and evaluated. The parameters were used for calculation of the plastic hinge location in the fastener on the timber side and the load-bearing capacity of the connections. For this, both the European Yield Model known from the Eurocode 5, as well as a model where the load-bearing capacity is determined from mechanical equilibrium of fasteners in deformed state were used. Double shear push-out tests were carried out to determine the load-carrying capacity of the connections. The calculated values are compared to the experimental results. The models showed similar results for the U-conn:s regarding both plastic hinge location and load-bearing capacity. For the nails, all models show good correlation with experimental values, however, the model derived from the deformed state predicted both the plastic hinge location and the load-carrying capacity more accurately.

A novel robust black-box fingerprinting scheme for deep classification neural networks

With the growing value and high training cost of deep neural network (DNN) models, there has been increasing research interest in deep neural network model watermarking. However, the existing literature lacks robust model watermarking methods with retrospective functionality, which hinders multi-user distribution and fine-grained copyright verification of deep neural network models. To achieve the model watermarking with retrospective functionality that is generally denoted the model fingerprinting, this study proposes a novel robust black-box fingerprinting scheme for deep classification neural networks, in which a fingerprint that is unique for each user is inserted in the given DNN. Specifically, a DWT-DCT-SVD-based embedding method is introduced to construct a high-quality poisoned image by spreading the fingerprint as a poisoned trigger to the entire image, which both leads to the traceability of the to-be-protected DNN and significantly enhances the stealthiness and anti-forgery of the embedded poisoned trigger. As the amplitude of the embedded fingerprint is rather small, it poses a challenge that makes the model hard to capture features of the embedded fingerprint during the training process. To tackle this challenge, this study then proposes a poisoned feature enhancement module, which groups classification categories of clean images without the fingerprint as one type while taking the poisoned images with the correct fingerprint and the poisoned adversarial images containing the other incorrect fingerprints as another two different types. This module, which can be removed during fingerprint verification, improves the reliability and fidelity of the model fingerprinting. Accordingly, an adversarial training strategy is designed, aiming at enhancing the effective training of classification neural network models that contain the fingerprint with small embedding strength. Combining these strategies then gives rise to the proposed scheme, which is denoted the UfNet for notational convenience. Extensive experimental simulation results demonstrate the excellent invisibility of the fingerprint in the poisoned image and the desirable verification of the fingerprint in case of even a one-bit difference. Additionally, it also exhibits robustness against the related attacks such as STRIP, Neural Cleanse, Fine-Pruning, and GradCAM, which outperforms the state-of-the-arts (SOTAs). Furthermore, the proposed UfNet is robust to collusion attacks and frame the innocent. These results show the feasibility and effectiveness of the introduced model fingerprinting framework and the proposed implementation methodology in this paper.

Multiple nonlinearities in the process of embedding and pulling out of nails: A numerical approach

This paper presents a numerical approach to analyze the embedding and withdrawal process of nails. Multiple nonlinearities are considered: law of orthotropic large strain elasto-plastic materials and contact with friction. The Uzawa algorithm based on the bi-potential method is used to describe the nonlinearity of boundary conditions. A return mapping algorithm is applied to deal with the elasto-plastic constitutive laws. On the basis of the Updated Lagrangian formulation, we adopt a Rotationally Neutralized Objective hypothesis to describe the geometrically nonlinear behavior. The Newton–Raphson iterative method is employed to solve the nonlinear equation. The simulation results show the deformation and the stress distribution of wood in the process of the nail embedding and withdrawal. The effectiveness and accuracy of the numerical scheme are proved by comparing with the existing engineering experiments. The influence of friction coefficients on the nail embedding and withdrawal strength is also explored.

Influence of test methodology on the characterization of the parallel-to-grain timber embedment strength and foundation modulus of dowels

ABSTRACT A reliable determination of the embedment strength and foundation modulus of timber elements is critical for the design and safety assessment of joints in timber structures. However, the existence of various test configurations for characterising the embedding properties of large diameter steel fasteners in timber elements poses challenges in directly comparing and utilising available test data. This paper aims to provide an insight into the influence of embedment property test methods, comparing experimental results from different test setups within the guidelines of the EN 383 and ASTM D 5764-97a standards for European softwood species, Scots pine wood (Pinus sylvestris) and Norway spruce (Picea abies). In addition to the test guidelines, the thickness of the specimen and the application of the load was evaluated within the protocols. A comprehensive statistical analysis was performed to identify statistically significant differences between the groups evaluated. The results of the analysis revealed disagreement between the standards in the evaluation of the strength of the embedding, highlighting the potential bias inserted by the experimental setup and protocol. Furthermore, it was proven that the thickness of the specimens influences both the embedding strength and the foundation modulus of the wood species tested. Finally, no distinctions were observed between tensile and compressive loading within the guidelines of the EN 383 standard.

Embedment strength and foundation modulus at an angle to the grain for fully threaded bolts

ABSTRACT As the main parameter, embedment strength is used to assess the load-carrying capacity of timber joints with dowel-type fasteners and foundation modulus is used to assess their stiffness. The embedment properties, i.e. embedment strength and foundation modulus, are system properties, which depend not only upon the material properties but also upon the load-to-grain angle and surface conditions of fasteners. The formulae for the embedment strengths of dowels and bolts in the design standards of timber structures are almost exclusively derived from the embedment tests on smooth dowels, with the surface conditions different from those of alternative fully threaded bolts in timber structures. Moreover, there are hardly any formulae for determining the foundation modulus. In this study, the embedment tests at different angles to the grain were carried out using the fully threaded bolts with different diameters in oversized holes as they would be used in practice. The load-to-grain angle had statistically significant effects on these embedment properties. The empirical formulae for the embedment strength and foundation modulus were derived. The embedment tests on fully threaded bolts and smooth dowels in tight-fitting holes were also performed to explore the effects of bolt hole sizes and surface conditions.

Impact of Acetylation on the Behaviour of Single-Dowel Timber Connections

This paper presents an experimental study where the mechanical behaviour of single-dowel timber connections made of acetylated Scots pine is compared with the behaviour of connections made from untreated Scots pine. The main aim was to evaluate the influence of the acetylation on the connection brittleness and also to compare the experimental results to the design provisions of the current European structural timber code, Eurocode 5 (EC5). The experiments included embedment tests and tests with connections loaded parallel and perpendicular to the grain, and, for the latter tests, applying different end and edge distances. The acetylated wood showed a 2% increase in density and a 31% increase in embedment strength compared to the untreated wood. For tests on connections loaded parallel to the grain, all specimens made from acetylated wood failed in a brittle manner, while the connections made from untreated wood and complying with minimum end distance of the EC5 design provisions failed due to embedment failure followed by splitting involving cracking along the grain. The connections made of acetylated wood showed a 13–15% higher capacity than the corresponding specimens made from untreated wood. Thus, to fully utilize the potential of the increased embedment strength parallel to the grain, it is concluded that reinforcement of the joint, e.g., by self-tapping screws or externally applied sheet reinforcement would be necessary if the minimum end distances of EC5 are applied. The current design provisions for loading perpendicular to the grain overestimated the capacities severely with predicted characteristic values being 20–50% higher than mean values from tests for the recommended minimum edge distances. Finally, it was found that the splitting capacity in loading perpendicular to the grain was 10–18% lower for the specimens made from acetylated wood compared to the untreated wood.

Color Image Watermarking Using Adaptive Tunable Q Wavelet, Finite Ridge Let, and Arnold Transforms with TSVD

In recent years, researchers have been working to integrate watermarking practices anchored in wavelet transform (WT) coefficients and Singular value decomposition (SVD) to recover the robustness of watermarked images in opposition to geometric and non-geometric attacks. Conversely, the watermark embedding, can initiate vagueness in the false-positive problem (FPP). To address this issue, Flamingo search algorithm-based Color Image Watermarking using Adaptive Tunable Q wavelet, Finite ridge let, Arnold transforms with Truncated Singular value decomposition (ATQWT-FRT-AT-TSVD-FSOA-CIW) is proposed. The proposed ATQWT-FRTAT-TSVD-FSOA-CIW method is separated into two stages: (i) watermark embedding and (ii) watermark extraction. The scaling factor is optimized with the help of the Flamingo search algorithm, which provides higher robustness, higher embedding strength, resistance to a false-positive problem, and better quality of the watermarking scheme. The proposed ATQWT-FRT-AT-TSVD-FSOA-CIW method is simulated in MATLAB tool. The performance of the proposed ATQWT-FRT-AT-TSVD-FSOA-CIW design is estimated under several performance metrics, such as Peak signal to noise ratio (PSNR), Structural similarity index measurement (SSIM), and Normalized cross-correlation (NCC). The performance of the proposed ATQWT-FRT-AT-TSVD-FSOA-CIW method attains 35%, 22.43%, 21.157%, 13.96%, and 7.186% higher PSNR compared to the existing designs, like secure with robust color image watermarking utilizing nature-inspired intelligence (DWT-SVD-ABCOA-CIW), the Artificial bee colony basis perceptually tuned blind color image watermarking in the hybrid LWT-DCT domain (LWT-DCT-ABCOA-CIW), the DWT-SVD based adaptive color multi-watermarking scheme for copyright protection under AMEF and PSO-GWO (DWT-SVD-Hybrid PSO-GWOA-CIW), Robust image watermarking based on fractional Krawtchouk transform with optimization (FrKT-Hybrid FO-CSOA-CIW), and the optimized semi-blind scheme for color image watermarking (DWTFRT-AT-SVD-PSO-CIW).

Multimodal Fusion With Optimized Embedding Strength for Consumptive Medical Image Protection

Multimodal Fusion With Optimized Embedding Strength for Consumptive Medical Image Protection

Experimental study on longitudinal embedding performance of bolted engineered bamboo connections

This paper summarizes the experimental results of longitudinal embedding tests of bolted engineered bamboo connections. Six groups of bolted engineered bamboo connection specimens were prepared and tested monotonically up to failure. Test variables were bolt diameter and type of engineered bamboo: laminated bamboo or bamboo scrimber. Connection stiffness and embedding strength of the specimens of bamboo scrimber were greater than those of the specimens of the laminated bamboo – reflecting the difference in material density. Embedding strength of the laminated bamboo was insensitive to bolt diameter whereas embedding strength of the bamboo scrimber decreased with an increase of the bolt diameter. It was shown that in order to accurately model bolt embedding into bamboo, a foundation model is required. Reduction factors for strength and modulus for such a model were calibrated using experimental results. A parametric study addressing friction and foundation model size is presented.

Feasibility of using Uruguayan Eucalyptus for sustainable timber connections

Uruguayan fast-growing Eucalyptus grandis has emerged as a promising sustainable material for structural building components. However, traditional methods for joining timber components, such as beams and columns, typically rely on metallic fasteners, and mass timber products like glued laminated timber often use adhesives. In response to growing environmental concerns over the past five years, researchers have explored the substitution of petroleum-based adhesives and metallic fasteners with wooden pegs or dowels. To create effective connections, these dowels must be fashioned from denser wood than the lamellae. The embedment strength of both woods with varying densities is a critical factor in the design of timber connections using fasteners. This paper presents the results of a structural performance analysis of fully eucalyptus dowelled connections, constructed by combining Eucalyptus grandis (? = 473.2 kg|m3, COV = 13.2 %) for the lamellae and Eucalyptus tereticornis (? = 1042.4 kg|m^3, COV = 3.2 %) for the dowel. The bending strength of dowels was determined for specimens with 15- and 18-mm diameters, yielding 173.0 MPa (COV = 10.7 %) and 184.0 MPa (COV = 8.3 %), respectively. Four series of connections were tested, varying the fiber orientation of the lamellae (longitudinal L, transversal T) and dowel diameter (15, 18), following the half-hole method described in ASTM D 5764 standard. The embedment strength values obtained for the connections are as follows: L-15 = 31.3 MPa (COV = 9.3 %), T-15 = 11.2 MPa (COV = 28.7 %), L-18 = 29.6 MPa (COV = 10.1 %), T-18 = 16.7 MPa (COV = 25.7 %). These results reveal that the longitudinal application of the load, aligned with the fiber orientation, exhibits approximately twice the strength of the transversal orientation, with less variability in the outcomes. Furthermore, a strong correlation with the lamellae density was observed, while no significant dependency on the dowel diameter was found. These findings represent a crucial initial step in characterizing and designing glue-free wood-wood connections. The utilization of different eucalyptus species presents a promising path for the development of connections that promote more sustainable timber construction systems.

Effect of Wood Densification and GFRP Reinforcement on the Embedment Strength of Poplar CLT

Embedment strength is an important factor in the design and performance of connections in timber structures. This study assesses the embedment strength of lag screws in three-ply cross-laminated timber (CLT) composed of densified poplar wood with densification ratios of 25% and 50%, under both longitudinal (L) and transverse (T) loading conditions. The embedment strength was thereafter compared with that of CLT reinforced with glass-fiber-reinforced polymer (GFRP). The experimental data was compared with results obtained using different models for calculating embedment strength. The findings indicated that the embedment strength of CLT specimens made of densified wood and GFRP was significantly greater than that of control specimens. CLT samples loaded in the L direction showed higher embedment strength compared to those in the T direction. In addition, 50% densification had the best performance, followed by 25% densification and GFRP reinforcement. Modelling using the NDS formula yielded the highest accuracy (mean absolute percentage error = 10.31%), followed by the Ubel and Blub (MAPE = 21%), Kennedy (MAPE = 28.86%), CSA (MAPE = 32.68%), and Dong (MAPE = 40.07%) equations. Overall, densification can be considered as an alternative to GFRP reinforcement in order to increase the embedment strength in CLT.

Adaptor: Improving the Robustness and Imperceptibility of Watermarking by the Adaptive Strength Factor

In watermarking, the watermark embedding strength is crucial, and the introduction of the strength factor can adjust the trade-off between the quality of the encoded image and the accuracy of the recovered message, thus enabling good imperceptibility and robustness of the encoded image. In traditional watermarking methods, the strength factor is selected in relation to the cover image, and based on different images, different strength factors are manually selected or algorithmically derived to adjust the visual effect of the watermarked image. However, due to the subjectivity and inflexibility of traditional algorithms, they can not achieve the effect of adaptive adjustment of watermarked images. Recently, watermarking methods combined with deep learning have gradually occupied the mainstream of this field. In the testing stage, to balance the overall robustness and imperceptibility, the strength factor is no longer selected based on the cover image as in traditional methods. Instead, it is set to a universal value. Therefore, the watermarking method based on deep learning is still in the primary stage of the trial-and-error method. To solve the subjectivity of the hand-designed embedding strength algorithm of the traditional watermarking methods and the low elasticity of the strength factor of the learning method so as to realize the adaptive embedding of watermarks, we propose an adaptive watermarking method with separate training. The proposed method adds a new component, the Adaptor, compared to other frameworks. The Adaptor can adaptively select strength factors to control the embedding strength of the watermark relying on the cover image and secret message. A two-stage training method is used to maintain the stability of the training and to achieve the best results for each component. With the results obtained from our experiments, our proposed method can find the appropriate strength factor and optimize it, resulting in improved robustness and imperceptibility of the watermark. The proposed method shows better results compared to the current state-of-the-art algorithms.

A two-stage robust reversible watermarking using polar harmonic transform for high robustness and capacity

Designing the robust reversible watermarking (RRW) with high robustness and capacity while maintaining invisibility and reversibility is a challenging problem. To alleviate this problem, we present a two-stage RRW scheme using the polar harmonic transform (PHT). In the first stage, it selects robust PHT moments for robust watermarking, develops an adaptive normalization strategy, and constructs a robust watermark embedding method. Specifically, the PHT moments with high robustness are first chosen via experimental simulations. The adaptive normalization strategy is then developed to adjust the robust watermark's embedding strength for the selected moments, aiming at achieving higher robustness. By adjusting quantized errors to integers, the standard distortion-compensated quantization index modulation (DC-QIM) is optimized. And the optimized DC-QIM for embedding robust watermark facilitates decreasing the number of bits of quantized errors representation and thus increasing the embedding capacity. In the second stage, the reversible watermarking is conducted to achieve reversibility in the absence of attacks. That is, distortions caused by the robust watermark embedding are inserted by a reversible watermarking technique. The proposed scheme performs better than state-of-the-arts in terms of bit error rate (BER) performance, according to extensive experimental results. Our code is available at https://github.com/yichao-tang/PHT-RRW.

Experimental investigation on embedment strength of bamboo-based composite prepared with the inorganic adhesive

Bamboo based composite materials have become an effective alternative for the green development in the field of civil engineering. This paper proposed a novel bamboo based composite prepared with the inorganic adhesive by adopting the cold pressing and molding technology, which is named “inorganic-bonded bamboo composite” and abbreviated as InorgBam, as well as the embedment strength of the novel bamboo-based composite was examined. The half-hole test method was adopted to investigate the embedment strength considering the influence factors of nominal dowel diameters and load-to-grain directions. Results indicated that the characteristic embedment strength parallel-to-grain of the InorgBam was unrelated to the dowel diameter when the nominal dowel diameter varies from 10 mm to 20 mm. However, the characteristic embedment strength perpendicular-to-grain declined linearly with the increasing nominal dowel diameters in the range of 10 mm–18 mm, and then remains constant within the range of 18 mm–25 mm in diameter. Hankinson formula for characteristic embedment strength at any angle was found in well agreement with experimental results of the InorgBam. The prediction formula of characteristic embedment strength for the novel InorgBam was corresponding proposed.

Improving robustness of spatial networks via reinforced nodes

Many real-world networks are embedded in space, and their resilience in the presence of reinforced nodes has not been studied. In this paper, we use a spatial network model with an exponential distribution of link length r and a characteristic length ζ to model such networks. We find that reinforced nodes can significantly increase the resilience of the networks, which varies with the strength of spatial embedding. We also study different reinforced node distribution strategies for improving the network's resilience. Interestingly, we find that the best strategy is highly dependent on the expected magnitude of failures which we analyze using percolation theory. Finally, we show that the reinforced nodes are analogous to an external field in the percolation phase transition and that their critical exponents satisfy Widom's relation.