The gradient descent algorithm has become the standard algorithm for computing the extreme values of functions, but for multivariate functions this algorithm is mostly ineffective. This is because the convergence rate of each element is inconsistent in most cases. However, in this paper, we found that the gradient sign is a self-optimizing operator, ensuring that the convergence rate is consistent across all elements. This also explains, from an optimization perspective, the success of the Fast Gradient Sign Method (FGSM) in generating adversarial samples that are indistinguishable from the normal input, but can easily fool neural networks. We also found that the fractional order gradient is also self-optimizing, and that the convergence speed of this algorithm can be controlled by adjusting the order of the gradient. Experiments suggest that this algorithm not only generates adversarial samples faster than other algorithms, but that a single source image can generate many such samples. This algorithm is also more effective than others at generating adversarial samples from simple images.

Atmospheric correction is a necessary step in image processing data and spectra recorded by spaceborne sensors for pure cloudless atmosphere, primarily in the visible and near-IR spectral range. We have present a fast and sufficiently accurate method of atmospheric correction based on the proposed analytical solutions describing with high accuracy the spectrum of outgoing radiation at the top boundary of the cloudless atmosphere. This technique includes the model of the atmosphere and its optical parameters that are important in terms of radiation transfer. The solution of the inverse problem for finding unknown parameters of the model is carried out by the method of non-linear least squares (Levenberg-Marquardt algorithm) for an individual selected pixel of the image (its spectrum), taking into account the adjacency effects. Using the found parameters of the atmosphere and the average surface albedo, assuming homogeneity of the atmosphere within a certain area of the hyperspectral image (or the whole frame), the spectral albedo at the Earth's surface is calculated for all other pixels. It is essential that the procedure of the numerical simulation with non-linear least squares of the direct transfer problem is based on using analytical solutions, which provides a very short calculation time of the atmospheric parameters (seconds or less) and the ability to perform atmospheric correction "on-fly." Testing methods of atmospheric correction was performed using the synthetic outgoing radiation spectra at the top of the atmosphere (TOA), obtained by numerical simulation in the LibRadTran code, as well as spectra of real space images of the Hyperion hyperspectrometer. A comparison with the results of atmospheric correction in module FLAASH of ENVI package has been performed. Finally, to validate our data obtained by the SHARK method, a comparative analysis with ground-based measurements of Radiometric Calibration Network (RadCalNet) was carried out.

Traffic sign detection plays a pivotal role in autonomous driving systems. The intricacy of the detection model necessitates high-performance hardware. Real-world traffic environments exhibit considerable variability and diversity, posing challenges for effective feature extraction by the model. Therefore, it is imperative to develop a detection model that is not only highly accurate but also lightweight. In this paper, we proposed YOLO-ADual, a novel lightweight model. Our method leverages the C3Dual and Adown lightweight modules as replacements for CPS and CBL modules in YOLOv5. The Adown module effectively mitigates feature loss during downsampling while reducing computational costs. Meanwhile, C3Dual optimizes the processing power for kernel feature extraction, enhancing computation efficiency while preserving network depth and feature extraction capability. Furthermore, the inclusion of the CBAM module enables the network to focus on salient information within the image, thus augmenting its feature representation capability. Our proposed algorithm achieves a [email protected] of 70.1% while significantly reducing the number of parameters and computational requirements to 51.83% and 64.73% of the original model, respectively. Compared to various lightweight models, our approach demonstrates competitive performance in terms of both computational efficiency and accuracy.

In the Versatile Video Coding (VVC) standard, affine motion models have been applied to enhance the resolution of complex motion patterns. However, due to the high computational complexity involved in affine motion estimation, real-time video processing applications face significant challenges. This paper focuses on optimizing affine motion estimation algorithms in the VVC environment and proposes a fast gradient iterative algorithm based on edge detection for efficient computation. Firstly, we establish judging conditions during the construction of affine motion candidate lists to streamline the redundant judging process. Secondly, we employ the Canny edge detection method for gradient assessment in the affine motion estimation process, thereby enhancing the iteration speed of affine motion vectors. Experimental results demonstrate that our affine motion estimation algorithm reduces encoding time by approximately 15%-35% while maintaining video bitrate and quality.

In this study, we evaluated and compared several major carbon sequestration methods based on the first principles, namely, energy use efficiency and elemental stoichiometry. This study suggested that, at present, wood burial is the only feasible carbon sequestration method because it can be implemented immediately on a large scale, is cost-effective and efficient, has a long sequestration time, has low technical requirements, and has a relatively low impact on agriculture.

It has remained a hard nut for years to segment sonar images, most of which are noisy images with inevitable blur after noise reduction. For the purpose of solutions to this problem, a fast segmentation algorithm is proposed on the basis of the gray value characteristics of sonar images. This algorithm is endowed with the advantage in no need of segmentation thresholds to be calculated. To realize this goal, it follows the undermentioned steps: first, calculate the gray matrix of the fuzzy image background. After adjusting the gray value, segment the region into the background region, buffer region and target regions. After filtering, reset the pixels with gray value lower than 255 to binarize images and eliminate most artifacts. Finally, remove the remaining noise from images by means of morphological image processing. The simulation results of several sonar images show that the algorithm can segment the fuzzy sonar image quickly and effectively, with no problem of incomplete image target shape. Thus, the stable and feasible method is testified.

Some medical professional organizations have advocated for including the menstrual cycle as a vital sign in adolescence, but not in adulthood. However, documenting menstrual cycle patterns is not routine clinical or research practice. Vital signs are used to predict health outcomes, indicate needed treatment, and monitor a clinical course. They can help identify pathologies, affirm wellness, and are responsive to exposures. Here we review the scientific evidence showing how the menstrual cycle meets these criteria and should therefore be treated as a vital sign. Using key words and controlled vocabulary terms, we carried out multiple literature searches, prioritizing the inclusion of systematic reviews, meta-analyses, and clinical practice guidelines. This review describes how the menstrual cycle is a health indicator, can cyclically impact health conditions, and its associations with long-term post-menopausal health outcomes. We review exposures influencing the menstrual cycle, evidence underlying its use to optimize wellness, and available tools for documenting cycles. Supplementary materials include patient handouts on menstrual cycle tracking, and an index of related clinical practice guidelines and reviews by subject. The menstrual cycle is a vital sign from menarche through menopause, an underutilized but powerful tool for understanding gynecological and general health.

We propose a computer-generated hologram method for near-eye virtual reality and augmented reality 3D display. A 3D object located near the holographic plane is projected onto a projection plane to obtain a plurality of projected images with different angles. The hologram is calculated by superposition of projected image convolution with a specific PSF through the interference of reference wave with object wave. Holographic 3D display systems with LED as illumination, 4f optical filtering system and lens as eyepiece for near-eye VR display and HOE as combiner for near-eye AR display are designed. The results show that the proposed method is about 38 times faster than the conventional point cloud method. The proposed method is flexible to produce speckle noise free high-quality VR and AR images with efficient focus and defocus capabilities.

Accurate and efficient traffic incident detection methods can effectively alleviate traffic congestion caused by traffic incidents, prevent secondary accidents and improve the safety of urban road traffic.Aiming at the problems that the traditional machine learning event detection method cannot fully extract the parameter characteristics of traffic flow and is not suitable formulti-dimensional and non-linear massive data, we propose a new traffic event detection method(CNN-XGBoost).This method combines the respective advantages of Convolution Neural Network(CNN) and Extreme Gradation Boosting (XGBoost). Firstly, we preprocessed the original freeway traffic incident detection data set by constructing initial variable set, data normalization, data balance processing and dimension reorganization. Secondly,we use CNN network to automatically extract the deep features of event detection data, and use XGBoost as a classifier to classify the extracted features for expressway traffic event detection.Finally, we use the data set of Hangzhou expressway microwave detector in China to carry out simulation experiments on CNN-XGBoost. The experimental results show that compared with XGBoost, CNN, Support Vector Machine (SVM) and Gradient Boosting Decision Tree (GBDT) and other methods, CNN-XGBoost method can effectively improve the accuracy of expressway traffic event detection and has better generalization ability.

A dynamic impact test was performed on as-rolled AZ31 alloy with gradient microstructure under various strains. The microstructural evolution and mechanical properties were systematically investigated. As the strain rate gradually increases, an increasing number of twins are formed, facilitating dynamic recrystallization (DRX), and the mechanical properties are also gradually improved as well. The microstructure becomes heterogeneous at higher strain rates, but the peak stress decreases. The impact process results in significantly higher performance due to microstructural refinement, work hardening by dislocations and precipitates. In addition, both the adiabatic shear band and the adjacent crack experience a temperature rise that exceeds the recrystallization temperature of the alloy. This observation also explains the presence of ultrafine recrystallized grains within the adiabatic shear band and the appearance of molten metal around the crack.

To solve large scale linear equations involved in the Fast Multipole Boundary Element Method (FM-BEM) efficiently, an iterative method named the generalized minimal residual method (GMRES)(m)algorithm with Variable Restart Parameter (VRP-GMRES(m) algorithm) is proposed. By properly changing a variable restart parameter for the GMRES(m) algorithm, the iteration stagnation problem resulting from improper selection of the parameter is resolved efficiently. Based on the framework of the VRP-GMRES(m) algorithm and the relevant properties of generalized inverse matrix, the projection of the error vector r_m+1 on r_m is deduced. The result proves that the proposed algorithm is not only rapidly convergent but also highly accurate. Numerical experiments further show that the new algorithm can significantly improve the computational efficiency and accuracy. Its superiorities will be much more remarkable when it is used to solve larger scale problems. Therefore, it has extensive prospects in the FM-BEM field and other scientific and engineering computing.

Airborne distributed position and orientation system (ADPOS), which integrates multi-inertia measurement units(IMUs), data-processing computer and Global Navigation Satellite System (GNSS), serves as a key sensor in new higher-resolution airborne remote sensing, such as array SAR, multinode imaging loads and so on. ADPOS can provide multinode high accuracy spatio-temporal reference information to realize multinode motion compensation with the various state estimation methods such as Central Difference Kalman Filtering (CDKF), modifed CDKF. However, these methods focus on noise optimization based on known nonlinear models and its linear minimum variance estimation criterion, which cannot gain better estimation performance for high-dimensional nonlinear systems. For this reason, the variational optimization process based on conjugate gradient algorithm is proposed for the estimation of filtering update step, which is more efficient than linear minimum variance estimation criterion. Therefore, aiming at ADPOS with high-dimensional and nonlinearity, Central Difference Variational Filtering (CDVF) using variational optimization based on conjugate gradient algorithm for mean correction in the filtering update stage is devoted to advancing the estimation accuracy of multinode motion parameters. A real ADPOS flight test has been conducted, the experimental results show that the accuracy of the slave motion parameters has a distinct progress than the up to date CDKF, and the compensation performance is distinct.

C-type lectin CD209/DC-SIGN and CD209L/L-SIGN proteins are distinct cell adhesion and pathogen recognition receptors that mediate cellular interactions and recognize a wide range of pathogens including, viruses such as SARS, SARS-CoV-2, bacteria, fungi and parasites. Pathogens exploit CD209L family proteins to promote infection and evade the immune recognition system. CD209L and CD209 are widely expressed in SARS-CoV-2 target organs and can contribute to infection and pathogenesis. CD209L family receptors are highly susceptible to alternative splicing and genomic polymorphism, which may influence virus tropism and transmission in vivo. The carbohydrate-recognition domain (CRD) and the neck/repeat region represent the key features of CD209L family proteins that are also central to facilitating cellular ligand interactions and pathogen recognition. While, the neck/repeat region is involved in oligomeric dimerization, the CRD recognizes the mannose containing structures present on specific glycoproteins such as those found on the SARS-CoV-2 spike protein. Considering the role of CD209L and related proteins in diverse pathogen recognition, this review article discusses the recent advances in the cellular and biochemical characterization of CD209 and CD209L and their roles in viral uptake, which has important implications in understanding the host-pathogen interaction, the viral pathobiology and driving vaccine development of SARS-CoV-2.

This study investigates the impact of plasma gradient modification and gradient illumination on the optoelectronic properties of buckypaper (BP), a flexible and large-scale material composed of multi-walled carbon nanotubes (MWCNTs). The BP samples were subjected to argon ion plasma treatment at varying power levels and durations, thereby creating different carrier concentration gradients on the surface. The photovoltage and photocurrent responses of the samples were then measured under uniform full illumination and gradient illumination conditions. The findings revealed that both plasma gradient modification and gradient illumination significantly enhanced the optoelectronic performance of BP. Notably, the combined application of these two methods yielded superior results compared to the application of either method alone. Specifically, the optimal plasma power for improving BP was found to be 20W. Under conditions of plasma gradient modification and gradient illumination, a photovoltage of 267.76 μV was generated, which represents a 21.44 times increase, and a photocurrent of 15.69 μA, reflecting a 32.69 times enhancement. The mechanism underlying this optoelectronic effect can be attributed to the presence of π-bonds in the carbon atoms. These π-bonds are excited by photons, resulting in the generation of small voltages and currents. This study underscores the potential of BP as an optoelectronic material and introduces a novel approach to enhance its optoelectronic properties through plasma gradient modification and gradient illumination.

The attitude dynamics and orbit dynamics of large rigid space structures are coupled with each other under gravity gradient, which may affect the stability of the large rigid space structures. In this paper, the gravity gradient stability of a large rigid space structure is studied under both small and large disturbances. Based on the rigid body dynamics and orbit dynamics, an accurate dynamic model without any linearization of the large rigid space structure is established via the natural coordinate formulation (NCF), which is able to describe the large overall motions of the structures. By using the generalized-α algorithm, the gravity gradient stability of the large rigid space structure is simulated and analyzed via various examples, including the influence of large disturbance angles, the positions at the stabilization and unstabilization regions. Finally, the relationship between spinning stability and gravity gradient stability is also investigated via a large spinning space structure.

Gradient based methods are popularly used in training neural networks and can be broadly categorized into first and second order methods. Second order methods have shown to have better convergence compared to first order methods, especially in solving highly nonlinear problems. The BFGS quasi-Newton method is the most commonly studied second order method for neural network training. Recent methods have shown to speed up the convergence of the BFGS method using the Nesterov’s acclerated gradient and momentum terms. The SR1 quasi-Newton method though less commonly used in training neural networks, are known to have interesting properties and provide good Hessian approximations when used with a trust-region approach. Thus, this paper aims to investigate accelerating the Symmetric Rank-1 (SR1) quasi-Newton method with the Nesterov’s gradient for training neural networks and briefly discuss its convergence. The performance of the proposed method is evaluated on a function approximation and image classification problem.

One of the important issues being explored in Industry 4.0 is a collaborative mobile robots. This collaboration requires precise navigation systems, especially indoor navigation systems where GNSS (Global Navigation Satellite System) cannot be used. To enable precise localization of robots, different variations of navigation systems are being developed, mainly based on trilateration and triangulation methods. Triangulation systems are distinguished by the fact that they allow precise determination of an object’s orientation, which is important for mobile robots. An important feature of positioning systems is the frequency of position updates based on measurements. For most systems it is 10-20 Hz. In our work, we propose a high-speed 50 Hz positioning system based on the triangulation method with infrared transmitters and receivers. In addition, our system is completely static, i.e. it has no moving/rotating measurement sensors, which makes it more resistant to disturbances (caused by vibrations, wear and tear of components, etc.). In this paper we describe the principle of the system as well as its design. Finally, we present tests of the built system, which show centimeter positioning accuracy with a position update frequency of 50 Hz.

Boron carbide (B4C) powders with defined stoichiometry, high crystallinity, minimal impurity content, and a fine particle size are imperative for realizing the exceptional properties of this compound in advanced high-technology applications. Nevertheless, achieving the desired stoi-chiometry and particle size using traditional synthesis methods, which rely on prolonged, high-temperature processes, can be challenging. The primary objective of this study is to syn-thesize fine B4C powders characterized by high crystallinity and a sub-micron particle size, em-ploying a fast, and energy-efficient method. B4C powders are synthesized from elemental boron and carbon in a high-frequency induction heating furnace using the Electromagnetic Induction Synthesis (EMIS) method. The rapid heating rate achieved through contactless heating promotes the ignition and propagation of the exothermic chemical reaction between boron and carbon. Additionally, electromagnetic effects accelerate atomic diffusion, allowing the reaction to be completed in an exceptionally short timeframe. The grain size and crystallinity of B4C can be finely tuned by adjusting various process parameters, including the post-ignition holding temperature and the duration of heating. As a result, fine B4C powders can be synthesized in under 10 minutes. Moreover, these synthesized B4C powders exhibit high resistance to oxidation when exposed to air.

Single particle analysis with cryogenic electron microscopy (cryo-EM) can determine high resolution structures of macromolecular assemblies. To reconstruct final three-dimension(3D) density map, a reliable initial model is essential and ab-initio reconstruction usually can avoid model bias. The grand challenge for ab-initio reconstruction is to alignment two-dimension(2D) low signal-to-noise ratio(SNR) of particle images with random and unknown parameters. Here we formulate a fast ab initio reconstruction method(named AlignFM) based on Fourier Mellin transform and particle-swarm intelligence optimization(PSO) algorithm. It enables simultaneous 2D alignment and 3D refinement, which can fast generate a reliable 3D map. After experiments on several real datasets. AlignFM shows high performance on fast ab initio reconstruction with a small amount of data.

A challenge in building large-scale superconducting quantum processors is the precise control and manipulation of the qubit state. However, the crosstalk between the microwave control lines impedes the parallel execution of high-fidelity digital and analog quantum operations. Here, we propose and demonstrate a universal compensation protocol for calibrating the microwave signal crosstalk. We also introduce amplified error sequences to optimize the accuracy. Furthermore, we show a definitive improvement in parallel gate operations with crosstalk cancellation, demonstrating the technique's effectiveness. This work paves the way for superconducting hardware that features automated calibration of microwave crosstalk, leading to enhanced fidelities in multiqubit circuits.

Today, mechanical tracking systems have been downsized to allow them to be used in the field of airborne laser communications and in the military domain. Risley systems are used for this purpose, which work by directing a beam of light to a given target point, this procedure is commonly known as the inverse problem. In this paper, an analytical method, the geometric method, has been designed and developed to determine the beam steering in a Risley system and solve the inverse problem. The method focuses on different geometric shapes, like circumference or ellipse, that are described when the beam passes through the second prism. The accuracy and efficiency of the geometric method has been analysed and found to be faster than the two-step method. Furthermore, the geometric method has been implemented in an iterative process and an accuracy of up to 1 pm has been achieved. This high accuracy would allow the geometric method to be applied in fields such as lithography, stereolithography or 3D printers.

Natural polluted insulator surfaces are always coated with various kinds of soluble constituents, and those constituents affect flashover performance differentially. Currently this fact is not considered either in laboratory experiments or field pollution degree measurements, causing the existing insulation selection method to be deficient. In this paper a systematical research on insulator flashover voltage gradient correction involving types of soluble pollution constituents was presented. Taking typical type glass insulator as the sample, its flashover tests polluted by typical soluble chemicals NaCl, NaNO₃, KNO₃, NH₄NO₃, MgSO₄, Ca(NO₃)₂ and CaSO₄ were carried out. Then the flashover gradient correction was made combining the flashover performance of each soluble constituent, the ESDD contribution of the seven constituents, as well as the saturation performance of CaSO₄. The correction was well verified with the flashover test results of insulator polluted by three types of soluble mixture. Research results indicate that the flashover gradient correction method proposed by this paper performs well in reducing the calculating error. It is recommended to carry out component measurements and flashover gradient correction to better select outdoor insulation configuration.

In this work, we show that a late fusion approach to multi-modality in sign language recognition improves the overall ability of the model in comparison to the singular approaches of Computer Vision (88.14%) and Leap Motion data classification (72.73%). With a large synchronous dataset of 18 BSL gestures collected from multiple subjects, two deep neural networks are benchmarked and compared to derive a best topology for each. The Vision model is implemented by a CNN and optimised MLP and the Leap Motion model is implemented by an evolutionary optimised deep MLP topology search. Next, the two best networks are fused for synchronised processing which results in a better overall result (94.44%) since complementary features are learnt in addition to the original task. The hypothesis is further supported by application of the three models to a set of completely unseen data where a multi-modality approach achieves the best results relative to the single sensor method. When transfer learning with the weights trained via BSL, all three models outperform standard random weight distribution when classifying ASL, and the best model overall for ASL classification was the transfer learning multi-modality approach which scored 82.55% accuracy.

The Median American Sign Language Interpretation Software (ASL) Interpretation Software is a web application that is capable of interpreting American Sign Language in real-time, utilizing an internet connection and a primary web camera, complete with basic phrases and letters. Extensive use of Deep Learning and Neural Networks, specifically Convoluted Neural Networks, enables Median to interpret video inputs and generate accurate results directly displayed to the user in text format. The ultimate goal for Median is to have it act as a bridge between hearing people and members of the deaf community, allowing deaf people to communicate with non-signing people using American Sign Language. Furthermore, Median has been designed to benefit people who lack access to a human ASL Translator, as its format as a website allows it to be accessible anywhere at any time, giving increased availability over human interpreters. Median is designed to be a very versatile program with great potential for growth and expansion.

In this paper the application of recycled Li-ion batteries spent cathode (LIB-SC) combined with NaHCO₃/H₂O₂ system is presented for the first time in the literature as an alternative for degradation of potentially toxic organic molecules. The model pollutant choice was methylene blue molecule. The spent cathode composition corresponds to LiCoO₂, which was proved by the XRD and EDX. Regarding the decolorization of methylene blue solution, the addition of NaHCO₃ in comparison with only H₂O₂ reduces the complete decolorization time in 96%. This reduction occurs because the radical CO₃ is more stable than OH. In this way, the application the system proposed in this article is aimed at solving two major global problems: the disposal of cell phone batteries and the pollution of liquid effluents.

Abstract: The Micro-Electro-Mechanical System (MEMS) LiDAR point cloud in autonomous vehicles has a large deflection range, which results in slow registration speed and poor applicability. To maximize speed, an improved NDT (Normal Distribution Transform) method that integrates point cloud density features has been proposed. First, the point cloud is reduced using a modified voxel filter and a pass-through filter. Next, the ISS (Intrinsic Shape Signature) algorithm is utilized to analyze the point cloud features and extract keypoints, the 4PCS (4-Point Congruent Sets) algorithm is then employed to calculate the initial pose under the constraints of the keypoint set to complete the coarse registration. Finally, the self-adaptive segmentation model is constructed by using KD-tree to obtain the density features of keypoints, and the NDT algorithm is combined with this model to form an SSM-NDT algorithm, which is used for fine registration. Each algorithm was compared on the autonomous vehicle dataset PandaSet and actual collected data sets, the results show that the novel method increases the speed by at least 60% and takes into account good registration accuracy and strong anti-interference.

High-throughput research on bimetallic nanoparticles brought a vast overview of their characteristics and catalytic activities.¹ However, traditional bimetallic nano-allloy quantification methods serving catalyst preparations are not suitable for the fast and robust data output of high-throughput experiments. Hence, this work designed and validated a more fast, more neat, and reliable quantitative analysis using X-ray fluorescence spectroscopy for bimetallic nanoparticles synthesized by high-throughput setups. The results demonstrated this method to be accurate with high precision. Furthermore, the agreement to qualitative results from an energy-dispersive X-ray spectrometer equiped on a scanning electron microscope provided an option to quantify a bimetallic nano-alloy by just analyzing one of the two composing elements. Thus, this method is proved to be highly compatible with high-throughput experiments.

For many engineering applications it is sufficient to use the concept of simple materials. However, higher gradients of the kinematic variables are taken into account to model materials with internal length scales as well as to describe localization effects using gradient theories in finite plasticity or fluid mechanics. In many approaches length scale parameter have been introduced that are related to a specific micro structure. An alternative approach is possible, if a thermodynamically consistent framework is used for material modelling as shown in the present contribution. However, even if sophisticated and thermodynamically consistent material models can be established there are still not yet standard experiments to determine higher order material constants. In order to contribute to this ongoing discussion system identification based on the method of self-adaptive filtering is proposed in this paper. To evaluate the effectiveness of this approach it has been applied to second-order gradient materials considering longitudinal vibrations. Based on thermodynamically consistent models that have been solved numerically it has been possible to prove that system identification based on self-adaptive filtering can be used effectively for both narrow-band and broadband signals in the field of second-order gradient materials. It has also been found that the differences identified for simple materials and gradient materials allow for condition monitoring and detection of gradient effects in the material behavior.

Sign language recognition technology can help people with hearing impairments to communicate with those who are hearing impaired. At present, with the rapid development of society, deep learning technology also provided certain technical support for sign language recognition work. In sign language recognition tasks, the use of traditional convolutional neural networks to extract spatio-temporal features from sign language videos suffers from insufficient feature extraction, resulting in low recognition rates. Nevertheless, video-based sign language datasets are very large and require a lot of computational resources for training and generalisation must be ensured, which poses a challenge for recognition. In this paper, we have presented a video-based sign language recognition method based on resnet and lstm. As the number of network layers increases, the ResNet network can effectively solve the granularity explosion problem and obtain better time series features. We use the Resnet convolutional network as the backbone model. At the initialisation stage, we obtain sign language features using ResNet; then, the learned feature space is used as the input of LSTM network to obtain long sequence features. The experimental results show that the accuracy of the above model is better than the mainstream model, and it can effectively extract the spatio-temporal features in sign language videos and improve the recognition rate of sign language actions.

Sign language recognition is essential in hearing-impaired people’s communication. Sign language recognition is an important concern in computer vision and has been developed with rapid progress in image recognition technology. However, sign language recognition using a general monocular camera has problems with occlusion and recognition accuracy in sign language recognition. In this research, we aim to improve accuracy by using a 2-axis bending sensor as an aid in addition to image recognition. We aim to achieve higher recognition accuracy by acquiring hand keypoint information of sign language actions captured by a monocular RGB camera and adding sensor assist. To improve sign language recognition, we need to propose new AI models. In addition, the amount of dataset is small because it uses the original data set of our laboratory. To learn using sensor data and image data, we used MediaPipe, CNN, and BiLSTM to perform sign language recognition. MediaPipe is a method for estimating the skeleton of the hand and face provided by Google. In addition, CNN is a method that can learn spatial information, and BiLSTM can learn time series data. Combining the CNN and BiLSTM methods yields higher recognition accuracy. We will use these techniques to learn hand skeletal information and sensor data. Additionally, the 2-axis Bending sensor glove data support training AI model. Using these methods, we aim to improve the recognition accuracy of sign language recognition by combining sensor data and hand skeleton data. Our method performed better than using skeletal information, achieving 96.5% accuracy in Top-1.

The electric vehicle (EV) industry is experiencing rapid growth, accompanied by continuous advancements in charging infrastructure to satisfy the rising need for fast and reliable charging. Particularly, the focus has shifted towards EV fast charging (EVFC) and its impact on the power grid. This paper presents a comprehensive analysis and survey of scholarly literature and projects conducted in this field over the past decade. The findings highlight the rising demand for EVFC and the need to address its potential adverse effects on the power grid. However, there is a noticeable lack of clear research guidance regarding charging time. To tackle these challenges and pave the way for future solutions, extensive research on DC fast charging, ultra-fast charging, and the integration of vehicle-to-grid (V2G) systems is required. This review paper thoroughly investigates the development of fast charging technology for electric vehicles (EVs), including its advantages and comparative analyses from various perspectives. Furthermore, it delves into the advancements in DC fast charging infrastructure, emphasizing charging standards and control modes. The study also investigates different categories of battery chargers for both on-board and off-board charging. Additionally, it explores the classification of DC-DC converters in the context of DC fast charging stations, discusses control strategies for EV systems, identifies existing challenges, and outlines future trends in EV fast charging.

Quantum computing is an emerging field that have given a significant impact on optimization. Among the diverse quantum algorithm, the quantum gradient descent has become a prominent technique for solving unconstrained optimization problems. In this paper, we propose a quantum-spectral Polak-Ribi{\'e}re-Polyak (PRP) conjugate gradient approach. The technique is considered as a generalization of the spectral PRP method which employs a $q$-gradient that approximate the classical gradient with quadratically better dependence on the quantum variable $q$. Additionally, the proposed method reduces to the classical variant as quantum variable $q$ gets closer to $1$. The quantum search direction always satisfies the sufficient descent condition and does not depend on any line search. This approach is globally convergent with the standard Wolfe line search without any convexity assumption. Numerical experiments are conducted and compared with the existing approach to demonstrate the improvement of the proposed strategy.

The object of this paper is the bending analysis of isotropic rectangular Kirchhoff plates subjected to non-uniform heating (NUH) using the Fourier transform method. The bottom and top surfaces of the plate are assumed to have different changes in temperature, whereas the change in temperature of the mid-surface is zero. According to classical plate theory, the governing equation of the plate contains second derivatives of the NUH; these derivatives are zero by constant value of the NUH, which leads to its absence in the governing equation. This paper presented an approach by which Fourier sine transform was utilized to describe the NUH, while the double trigonometric series of Navier and the simple trigonometric series of Lévy were utilized to describe the deflection curve. Thus, the NUH appeared in the governing equation, which simplified the analysis. Rectangular plates simply supported along all edges were analyzed, bending moments, twisting moments, and deflections being determined. In addition, rectangular plates simply supported along two opposite edges were analyzed; the other edges having various support conditions (free, simply supported, and fixed).

This paper explores the capability of various machine learning algorithms, including Random Forests and advanced gradient boosting techniques such as XGBoost, LightGBM, and CatBoost, to predict customer churn in the telecommunications sector. For this analysis, a dataset available to the public was employed. The performance of these algorithms was assessed using recognized metrics, including Accuracy, Precision, Recall, F1-score, and the Receiver Operating Characteristic Area Under Curve (ROC AUC). These metrics were evaluated at different phases: subsequent to data preprocessing and feature selection; following the application of SMOTE and ADASYN sampling methods; and after conducting hyperparameter tuning on the data that had been adjusted by SMOTE and ADASYN. The outcomes underscore the notable efficiency of upsampling techniques such as SMOTE and ADASYN in addressing the imbalance inherent in customer churn prediction. Notably, the application of random grid search for hyperparameter optimization did not significantly alter the results, which remained comparatively unchanged. The algorithms' performance post- ADASYN application marginally surpassed that observed after SMOTE application. Remarkably, LightGBM achieved an exceptional F1-score of 89% and an ROC AUC of 95% subsequent to the ADASYN sampling technique. This underlines the effectiveness of advanced boosting algorithms and upsampling methods like SMOTE and ADASYN in navigating the complexities of imbalanced datasets and intricate feature interdependencies.

Questions：SAP in mountainous cities of China serve as crucial carriers for species and wild food, constituting potential green infrastructure. These patches should be recognized as locally specific resources. This study investigates the transformation characteristics of the landscape patterns of SAP, along with their elements, across urban-rural gradients, providing foundational support for the planning of conservation and utilization strategies. Location: Yongchuan area of Chongqing in southwest China. Methods: The study investigated urban-rural gradient-based transformation of SAP. A four-direction transect space was selected and divided into uniform sample squares. Fragstats 4.3 calculated the landscape index of SAP, and Spearman correlation coefficients in SPSS were used to evaluate the influence of landform and artificial factors on agroforestry patch transformation. Results: 1) Patch area (PA), Plaque density (PD), and CLUMPY of SAP systematically transform, showing increased SAP areas and higher connectivity with greater distance from the city, resulting in reduced patch fragmentation. However, Edge density (ED) and Largest plaque index (LPI) exhibit irregular transformation patterns. 2) With increased distance from the city, woodland patch area and connectivity increase, leading to decreased fragmentation, dispersal, and lower fractal dimensions. Conversely, Cultivated land follows the opposite trend. garden land patch area, fractal dimension, fragmentation, and dispersal all increase, while connectivity decreases. 3) Landscape pattern transformations of SPA are significantly influenced by artificial factors. Woodland patches are influenced similarly by landform and artificial factors. In contrast, cultivated land and garden land patches are more significantly impacted by artificial factors. Conclusions: SAP exhibit differentiated conservation and utilization measures across urban-rural gradients. Small patches located far from urban centers may benefit from systematic conservation measures to prevent further fragmentation and erosion in peri-urban areas.

(1) Background: Depressional wetlands are highly vulnerable to changes in temperature and rainfall, affecting littoral vegetation, thus HydroGeoMorphic (HGM) wetland types. This study aimed to assess the utility of Sentinel-2A vegetation indices in enhancing delineation and sub-classify depressional HGM types. (2) Methods: We sampled 84 pairs of four vegetation indices from 10x10m plots along 14 belt transects in eight depressional wetlands in Mpumalanga, South Africa. We evaluated vegetation index differences between and within wetlands, focusing on lit-toral gradients. (3) Results: Significant differences (Bonferroni p

In the era of Artificial Intelligence (AI), comprehending and responding to non-verbal communication is increasingly vital. This research extends AI's reach in bridging communication gaps, notably benefiting American Sign Language (ASL) and Taiwan Sign Language (TSL) communities. It focuses on employing various AI models, especially the Hierarchical Vision Transformer with Shifted Windows (Swin), for recognizing diverse sign language datasets. The study assesses Swin architecture's adaptability to different sign languages, aiming to create a universal platform for Unvoiced communities. Utilizing deep learning and transformer technologies, hybrid application prototypes have been developed for ASL-to-English translations and vice versa, with plans to expand this to multiple sign languages. The Swin models, trained on varied dataset sizes, show considerable accuracy, indicating their flexibility and effectiveness. This research underscores major advancements in sign language recognition and underlines a commitment to inclusive communication in the digital era. Future work will focus on enhancing these models and broadening their scope to include more sign languages, integrating multimodality and Large Language Models (LLMs), thereby fostering global inclusivity.

The movement of liquid droplets on the energy gradient surface has attracted extensive attention. Inspired by biological features in nature, such as the periodic spindle-shaped nodes in spider silks and conical-like barbs of cacti, the structure-property-function relationship of multifunctional gradient surfaces. In this study, a series of specific patterns are fabricated by 3D printing technology, followed by modification via the atmospheric pressure plasma treatment and liquid phase chemical deposition, resulting in enhancing the ability of water droplets of 5μL to travel 18.47mm on a horizontal plane and 22.75mm against gravity up to 20° tilting angle. Additionally, analysis techniques have been employed including contact angle analyzer, ESCA and laser confocal microscope to evaluate the sample performance. This work could further be applied to many applications related to microfluidic devices, drug delivery and water/fog collection.

Background: Microscopic examination of acid-fast stained sputum smears for detecting mycobacterial acid-fast bacilli (AFB) remains the most economical and readily available method for laboratory diagnosis of tuberculosis (TB). However, this conventional approach has limitations, including low sensitivity and labor-intensive procedures. Methods: An automated microscopy system incorporating artificial intelligence (AI) and machine learning for AFB identification was evaluated. The study was conducted at an Infectious Disease Hospital in Jiangsu Province, China, utilizing an intelligent microscope system (TB-Scan, Wellgen Medical, Kaohsiung). A total of 1,000 sputum smears were included in the analysis, with the system capturing digital microscopic images and employing an image recognition model to automatically identify and classify AFB. Referee technicians provided the gold standard for discrepant results. Results: The automated system demonstrated an overall accuracy of 95.00% (950/1,000), sensitivity of 91.24% (177/194), and specificity of 95.91% (773/806). Notably, the system identified 21 smears as positive that were previously reported as negative, with referee technicians confirming 17 of these as true positive and retracting the test results accordingly. Recalculating the performance, the accuracy increased to 96.70% (967/1,000), sensitivity to 91.94% (194/211), and specificity to 97.97% (773/789), with a false negative rate of 8.06% (17/211) and a false positive rate of 2.03% (16/789). Conclusions: The incorporation of AI and machine learning into an automated microscopy system demonstrated the potential to enhance the sensitivity and efficiency of AFB detection in sputum smears compared to conventional manual microscopy. This approach holds promise for widespread application in TB diagnostics and potentially other fields requiring labor-intensive microscopic examination.

In Mexico, the incorporation of deaf people into education has been lacking since only 14% of the deaf population in the age group between 3 and 29 years access education with the support of a hearing aid. Additionally, those who have been incorporated frequently face inappropriate educational strategies which poorly develop the use of Mexican Sign Language (MSL) and therefore academical success and opportunities for insertion in the workplace are difficult. This research explores a novel mexican sign language lexicon video dataset containing the dynamical gestures most frequently used by MSL. Each gesture consists of a set of different versions of videos under uncontrolled conditions. MX-ITESO-100 data set is composed of a lexicon of 100 gestures and 5,000 videos from three participants with different grammatical elements. Additionally, the data set is evaluated in a two-step neural network model with an accuracy greater than 99%. and thus serves as a benchmark for future training of machine learning models in computer vision systems. Finally, this research provides an inclusive environment within society and organizations in particular for people with hearing impairment.

As a new type of high-performance material, gradient structural steel is widely used in engineering fields due to its unique microstructure and excellent mechanical properties. For the prevalent fatigue failure problem, the rate of change of the local grain size gradients along the structure (referred to as the gradient rate) is a key parameter in the design of gradient structures, which significantly affects the fatigue performance of gradient structural steel. In this work, a method of ‘Voronoi primary + secondary modeling’ is adopted to successfully establish three typical high-strength steel models corresponding to the convex, linear, and concave type gradient rates for gradient structures, focusing on the stress-strain response and crack propagation in structural steel with different gradient rates under cyclic loading. It is found that the concave gradient rate structural model is dominated by finer grains with larger volume fraction, which is conducive to hindering fatigue crack propagation and has the longest fatigue life; the linear structure is the second one, and the convex structure is the shortest one. The simulation results in this work are consistent with the relevant experimental phenomena. Therefore, when regulating the gradient rate, priority should be given to increasing the volume fraction of fine grains and designing a gradient rate structure dominated by fine grains to improve the fatigue life of the material.

Quantifying the landscape pattern change can effectively demonstrate the ecological progresses and the consequences of urbanization. Based on remotely sensed land cover data in 1994, 2000, 2006 and a gradient analysis with landscape metrics at landscape- and class- level, we attempted to characterize the individual and entire landscape patterns of Shanghai metropolitan during the rapid urbanization. We highlighted that a roadscape transect approach that combined the buffer zone method and the transect-based approach was introduced to describe the urban-rural patterns of agricultural, residential, green, industrial, and public facilities land along the railway route. Our results of landscape metrics showed significant spatiotemporal patterns and gradient variations along the transect. The urban growth pattern in two time spans conform to the hypothesis for diffusion–coalescence processes, implying that the railway is adaptive as a gradient element to analyze the landscape patterns with urbanization. As the natural landscape was replaced by urban landscape gradually, the urban fringe expanded radically. The results also showed that the desakota region expanded its extent widely. Satellite towns witnessed the continual transformation from the predominantly rural landscape to peri-urban landscape. Furthermore, the gap between urban and rural areas remained large especially in public service. More reasonable urban plans and land use policies should push to make more of an effort to transition from the urban-rural separation to coordinated urban-rural development. This study is a meaningful trial in demonstrating a new form of urban–rural transects to study the landscape change of large cities from a strategic viewpoint. By combining gradient analysis with landscape metrics, we addressed the process of urbanization both spatially and temporally, and provided a more quantitative approach to urban studies.

In wideband radar systems, the performance of motion parameters estimation can significantly affect the performance of object detection and the quality of inverse synthetic aperture radar (ISAR) imaging. Although the traditional motion parameters estimation methods can reduce the range migration (RM) and Doppler frequency migration (DFM) effects in ISAR imaging, the computational complexity is high. In this paper, we propose a new fast non-searching motion parameters estimation method based on cross-correlation of adjacent echoes (CCAE) for wideband LFM signals. A cross-correlation operation is carried out for two adjacent echo signals, then the motion parameters can be calculated by estimating the frequency of the correlation result. The proposed CCAE method can be applied directly to the stretching system, which is commonly adopted in wideband radar systems. Simulational results demonstrate that the new method can achieve better estimation performances, with much lower computational cost, compared with existing methods. The experimental results on real radar data sets are also evaluated to verify the effectiveness and superiority of the proposed method compared to the state-of-the-art existing methods.

Transportation contributes significantly to greenhouse gas emissions, with electrification offering a promising solution. This study evaluates the degradation of State of Health (SOH) over 2.5 years for electrified buses under different charging protocols: DC Fast charging and Flash charging. The vehicles’ models, charging protocols, and financial implications are discussed. High-level simulations reveal that after 2.5 years, the Flash charging vehicle’s SOH reaches 88.93%, while the DC fast charging vehicle reaches 89.46%, showing a 0.47% difference. This slight variance is attributed to Flash charging vehicles avoiding full charge/discharge cycles, reducing battery aging impact. Considering this minimal SOH degradation difference, the choice between DC Fast charging and Flash charging primarily hinges on capital expenditure disparities. Flash charging, with its lower initial battery capacity and reduced capital costs, presents a compelling alternative for city busoperators seeking to electrify their fleets.

Fast-scan cyclic voltammetry (FSCV) is an electrochemical sensing technique that can be used for neurochemical sensing with high spatiotemporal resolution. Carbon fiber microelectrodes (CFMEs) are traditionally used as FSCV sensors. However, CFMEs are prone to electrochemical fouling caused by oxidative byproducts of repeated serotonin (5-HT) exposure, which makes them less suitable as a chronic 5-HT sensor. Our team is developing a boron-doped diamond microelectrode (BDDME) that has previously been shown to be relatively resistant to fouling caused by protein adsorption (biofouling). We sought to determine if the BDDME exhibits resistance to electrochemical fouling, which we explored on electrodes fabricated with either femtosecond laser cutting or physical cleaving. We recorded the oxidation current response after 25 repeated injections of 5-HT in a flow-injection-cell and compared the current drop from the first to the last injection. The 5-HT responses were compared to dopamine (DA), a neurochemical that is known to produce minimal fouling oxidative byproducts and have a stable repeated response. Physical cleaving of the BDDME yielded a reduction in fouling due to 5-HT compared to the CFME and the femtosecond laser cut BDDME. However, the femtosecond laser cut BDDME exhibited a large increase in sensitivity over the cleaved BDDME. An extended stability analysis was conducted for all device types following 5-HT fouling tests. This analysis demonstrated an improvement in the long-term stability of boron doped diamond over CFMEs, as well as a diminishing sensitivity of the laser cut BDDME over time. This work reports the electrochemical fouling performance of the BDDME when it is repeatedly exposed to DA or 5-HT, which informs the development of a chronic, diamond-based electrochemical sensor for long-term neurotransmitter measurements in vivo.

Global optimization is widely adopted nowadays in a variety of practical and scientific problems. In this context, a group of techniques that is widely used is that of evolutionary techniques. A relatively new evolutionary technique in this direction is that of Giant Armadillo Optimization, which is based on the hunting strategy of giant armadillos. In this paper, a number of modifications to this technique are proposed, such as the periodic application of a local minimization method as well as the use of modern termination techniques based on statistical observations. The proposed modifications have been tested on a wide - series test functions, available from the relevant literature and it was compared against other evolutionary methods.

The aim of this study is to develop a software solution for real-time recognition of sign language words using two arms. This will enable communication between hearing-impaired individuals and those who can hear. Several sign language recognition systems have been developed using different technologies, including cameras, armbands, and gloves. The system developed in this study utilizes surface electromyography (muscle activity) and inertial measurement unit (motion dynamics) data from both arms. Other methods often have drawbacks, such as high costs, low accuracy due to ambient light and obstacles, and complex hardware requirements, which have prevented their practical application. A software has been developed that can run on different operating systems using digital signal processing and machine learning methods specific to the study. For the test, we created a dataset of 80 words based on their frequency of use in daily life and performed a thorough feature extraction process. We tested the recognition performance using various classifiers and parameters and compared the results. The Random Forest algorithm was found to have the highest success rate with 99.875% accuracy, while the Naive Bayes algorithm had the lowest success rate with 87.625% accuracy. Feedback from a test group of 10 people indicated that the system is user-friendly, aesthetically appealing, and practically useful. The new system enables smoother communication for people with hearing disabilities and promises seamless integration into daily life without compromising user comfort or lifestyle quality.

The deaf-mute and individuals with hearing disabilities who communicate through sign language should not be overlooked or excluded. It is a social responsibility to strive for an inclusive world without inequality. Advances in technology offer various solutions to address the communication gap, one of which is the use of computer-based sign language recognition systems. This project proposes a Malaysian Sign Language recognition model utilizing MobileNetV1, a convolutional neural network-based algorithm, to interpret sign language in real-time settings through a website application using Streamlit. The outcome of the model test shows that it has achieved a recognition rate of 96 percent. Although the proposed model has demonstrated the ability to recognize sign language, it is suggested that further improvement can be achieved by enhancing the quality and diversity of the dataset. The implementation of a more comprehensive dataset would lead to improved performance and increased accuracy of the model. This project underscores the importance of considering and accommodating the needs of the deaf-mute and individuals with hearing disabilities and highlights the potential of technology to bridge communication barriers and achieve a more inclusive society.

Next-generation sequencing (NGS) has actualized human papillomavirus (HPV) virome profiling for in-depth investigation of viral evolution and pathogenesis. However, viral computational analysis remains a bottleneck due to semantic discrepancies between computational tools and curated reference genomes. To address this, we developed and tested automated workflows for HPV taxonomic profiling and visualization using a customized Papillomavirus database in CLC Microbial Genomics Module. HPV genomes from Papilloma Virus Episteme were customized and incorporated into CLC “ready-to-use” workflows for stepwise data processing to include: 1) Taxonomic Analysis, 2) Estimate Alpha/Beta Diversities, and 3) Map Reads to Reference. Low-grade (n = 95) and high-grade (n = 60) Pap smears were tested with ensuing collective runtimes: Taxonomic Analysis (36 min); Alpha/Beta Diversities (5 sec); Map Reads (45 min). Tabular output conversion to visualizations entailed 1-2 keystrokes. Biodiversity analysis between low- (LSIL) and high-grade squamous intraepithelial lesions (HSIL) revealed loss of species richness and gain of dominance by HPV-16 in HSIL. Integrating clinically relevant, taxonomized HPV reference genomes within automated workflows proved to be an ultra-fast method of virome profiling. The entire process named “HPV DeepSeq” provides a simple, accurate and practical means of NGS data analysis for a broad range of applications in viral research.

As technology advances, people are increasingly adopting IoT (Internet of Things) devices in smart cities. These devices are equipped with sensors that are connected everywhere, from energy infrastructure and transportation to telecommunications, health and human services, waste management, public safety, and more. People want to access sensor data anytime and everywhere to make better decisions and improve their lives. To achieve this, a sensor integration model has been developed where multiple sensors are integrated under one gateway to manage and share sensor data. However, effective sharing of sensor data requires data security, privacy, confidentiality, fairness, effective architectural design, and low communication and computation costs. These issues need to be addressed in the sensor integration model in smart cities. Existing storage models face challenges in managing sensor data effectively. To overcome these issues and hurdles, a data storage framework leveraging blockchain technology has been proposed within the sensor integration model. This framework ensures data security, privacy, confidentiality, and a fair and transparent financial model, all while minimizing communication and computation costs.

Poplars (Populus spp.) are of significant ecological and economic importance. Long-term breeding efforts were aimed mainly at obtaining fast-growing and productive plants and less considered resistance to pests. This study aimed to identify the patterns of susceptibility to Saperda carcharias (Linnaeus, 1758) (Coleoptera: Cerambycidae) among Populus hybrids and pure species, considering crossing combinations and some other traits. In our research, 35 clones of poplar species and hybrids of the Ukrainian and foreign selection from the Aigeiros, Tacamahaca, and Leucoides sections were tested in 2021 and 2023 in the plantation created in 2014 in Eastern Ukraine. S. carcharias preferred infesting clones with earlier foliage development, larger diameters, and height increments. Specifically, hybrids with American maternal lineage showed the highest susceptibility, whereas those with European maternal lineage or Asian origin displayed lower infestation levels or resistance, respectively. This variability suggests a complex relationship between genetic background and pest resistance, underlining the need for a deeper understanding of the factors influencing susceptibility to S. carcharias. Selecting native species clones or creating mixed clone plantations could enhance the resilience of poplar plantations to pest threats.

Timber-concrete composite (TCC) systems join the positive aspects of engineered wood products (good seismic behaviour, low thermal conductivity, environmental sustainability, good behaviour under fire if appropriately designed) with those of concrete (high thermal inertia, durability, excellent fire resistance). TCC facades are typically composed by an internal insulated timber-frame wall and an external concrete slab, separated by a ventilated air cavity. However, there is very limited knowledge concerning the performance of TCC facades, especially for what concerns their thermal behaviour. The present paper deals with the development and optimisation of a 2D CFD model for the analysis of TCC ventilated façades thermal behaviour. The model is calibrated and validated against the experimental data collected during the annual monitoring of a real TCC ventilated envelope in the north of Italy. Also, a new solver algorithm is developed to significantly speed up the simulation. The final model can be used for the time-efficient analysis and optimisation of the thermal performance of TCC ventilated facades, as well as other ventilated facades with external massive cladding, avoiding the expensive and time-consuming construction of mock-ups, or the use of comparably slow (conventional) CFD solvers that are less suitable for optimization studies.

Although Eulerian approaches are standard in computational acoustics, they are less effective for certain classes of problems like bubble acoustics and combustion noise. A different approach for solving acoustic problems is to compute with individual particles following particle motion. In this paper, a Lagrangian approach to model sound propagation in moving fluid is presented and implemented numerically, using three meshfree methods to solve the Lagrangian acoustic perturbation equations (LAPE) in the time domain. The LAPE split the fluid dynamic equations into a set of hydrodynamic equations for the motion of fluid particles and perturbation equations for the acoustic quantities corresponding to each fluid particle. Then, three meshfree methods, the smoothed particle hydrodynamics (SPH) method, the corrective smoothed particle (CSP) method, and the generalized finite difference (GFD) method, are introduced to solve the LAPE and the linearized LAPE (LLAPE). The SPH and CSP methods are widely used meshfree methods, while the GFD method based on the Taylor series expansion can be easily extended to higher orders. Applications to modeling sound propagation in steady or unsteady fluids in motion are outlined, treating a number of different cases in one and two space dimensions. A comparison of the LAPE and the LLAPE using the three meshfree methods is also presented. The Lagrangian approach shows good agreement with exact solutions. The comparison indicates that the CSP and GFD method exhibit convergence in cases with different background flow. The GFD method is more accurate, while the CSP method can handle higher Courant numbers.

In this article authors shows chosen problems of technical state diagnosis with the use of identification and technical diagnostics methods such as experimental modal analysis. Relations between methods of dynamic state evaluation and methods of technical state evaluation were indicated. Example modal analysis results illustrate the complexity of projecting dynamic state researches into diagnostic researches of state evaluation.

Increased anthropogenic pressure on forest resources leads to deforestation and forest degradation, significantly limiting the regeneration capacity of native woody species and consequently the restoration of miombo woodlands in anthropized habitats within the rural area of Lubumbashi. This study assessed the diversity and natural regeneration capacity of miombo species through floristic inventories in three different habitats (unexploited forests, degraded forests, and post-cultivation fallows). The results reveal that for the adult’s stratum unexploited and degraded forests exhibit higher dendrometric (density, mean square diameter, basal area) and floristic parameter (taxa, genera, families) values compared to post-cultivation fallows. Furthermore, the regeneration of miombo woody species is higher in degraded forests (21 taxa; 105 juveniles/plot). However, regarding the sapling’s stratum (1 cm ≤ dbh < 10 cm), the three habitats display nearly similar situations. Additionally, the floristic composition and diversity of unexploited and degraded forests show a significantly higher similarity (76.50%) among them compared to these habitats and the post-cultivation fallows (56.00%). These findings indicate that miombo woodlands have the potential to regenerate and maintain floristic diversity. To sustain this natural regeneration capacity of miombo woody species and promote the restoration of forest mass and its floristic diversity, it is imperative to determine the rotation period after habitat exploitation and regulate anthropogenic activities and late bush fires, particularly in anthropized habitats at the village level.

The paper describes the new method for image compression based on coordinated group signal transformation. The algorithm is a type of difference coding. Coordinated processing significantly simplifies the difference signal conversion scheme using a single group codec for all signals. The described method considers color channels as correlated signals of a multi-channel communication system. The specifics of the processed data required modification of the coordinated group processing algorithm. First, we changed how to calculate the difference signals to prevent data loss. Secondly, the group codec was supplemented with a neural network to improve the quality of reconstructed images. We considered the following types of neural networks: fully connected, recurrent, convolution, and convolution in the Fourier space. Based on the simulation results, it is best to use fully connected neural networks if the goal is to minimize processing delay time. These networks have a response time of 13 ms. On the other hand, if the priority is to improve quality in cases where delays are not critical, then convolution neural networks in the Fourier space should be used. These networks have a response time of 140 ms and are of significant interest.

In vertical coordinates, (x,y,z), the gradient Richardson number Ri=N^2⁄S^2 is the square of the ratio between buoyancy frequency N and vertical-shear amplitude S, where N^2=-(g⁄ρ) ∂ρ⁄∂z and S^2=(∂u⁄∂z)^2+ (∂v⁄∂z)^2, with ρ potential density, (u,v) the horizontal velocity components and g gravity acceleration, so that a decrease in stratification brings a decrease in dynamic stability and vice versa. In the isopycnic coordinate system, (x,y,ρ), however, Ri=M^2⁄〖S_ρ〗^2 , is the ratio between 〖M^2≡N〗^(-2) and the squared diapycnal shear 〖S_ρ〗^2=(ρ⁄g)^2 [(∂u⁄∂ρ)^2+ (∂v⁄∂ρ)^2 ], so a decrease in stratification leads to an increase in dynamic stability and vice versa. The apparently different role of stratification arises because S and S_ρ are related through the stratification itself, S_ρ=S⁄N^2 or, equivalently, τ=〖t_o〗^2⁄t_d , which is interpreted as a natural oscillation time τ≡S_ρ that equals the buoyancy or oscillation time t_o=N^(-1) normalized by the ratio t_o⁄t_d , where the deformation time is t_d=S^(-1). Here we follow simple arguments and use field data from three different situations (island shelf-break, Gulf Stream and Mediterranean outflow) to endorse the usefulness of the isopycnal approach. In particular, we define the reduced squared vertical σ^2= S^2-N^2 and reduced squared diapycnal shear 〖σ_ρ〗^2= 〖S_ρ〗^2-M^2, which are positive for unstable conditions and negative for stable conditions. It turns out that Ri and σ^2 remain highly variable for all stratification conditions; in contrast, the mean 〖σ_ρ〗^2 values display a univocal tendency, its magnitude decreasing with stratification. We propose 〖σ_ρ〗^2 and 〖S_ρ〗^2 to be good indexes for the occurrence of effective mixing under high stratification conditions.

Gradient-domain image processing is a technique where, instead of operating directly on the image pixel values, the gradient of the image is computed and processed. The resulting image is obtained by reintegrating the processed gradient. This is normally done by solving the Poisson equation, most oftenly by means of a finite difference implementation of the gradient descent method. However, this technique in some cases lead to severe haloing artefacts in the resulting image. To deal with this, local or anisotropic diffusion has been added as an ad-hoc modification of the Poisson equation. In this paper, we show that a version of anisotropic gradient-domain image processing can result from a more general variational formulation through the minimisation of a functional formulated in terms of the eigenvalues of the structure tensor of the differences between the processed gradient and the gradient of the original image. Example applications of linear and non-linear local contrast enhancement and colour image daltonisation illustrate the behaviour of the method.

The main focus of the in vitro toxicity assessment methods is on the viability of the cells, which is usually based on metabolism changes. Yet, when exposed to toxic substances, the cell triggers multiple signals in response. With this in mind, we have developed a promising cell-based toxicity method that observes various cell responses when exposed to toxic substances (either death, division or remaining viable). Based on the collective cell response, we observed and predicted the dynamics of the cell population to determine the toxicity of the toxicant. The method was tested with two different conformations: In the first conformation, we exposed a monoculture model of blood macrophages to UV light, hydrogen peroxide, nutrient deprivation, tetrabromobisphenol A, fatty acids and 5-fluorouracil. In the second, we exposed a coculture liver model consisting of hepatocytes, hepatic stellate cells, Kupffer cells and liver sinusoidal endothelial cells to rifampicin, ibuprofen and 5-fluorouracil. The method showed good accuracy compared to established toxicity assessment methods. In addition, this approach provided more representative information on the toxic effects of the compounds as it considers the different cellular responses induced by toxic agents.

This study addresses the challenge of treating skin cancer effectively, especially in visible areas like the face, where traditional methods may be undesirable. Current options, including surgery and radiation therapy, have limitations such as complexity, expense, and imprecise dose delivery. The proposed solution involves additively manufacturing personalized patches using diagnostic imaging data, offering a targeted and controlled dose of beta-emitting isotopes like Y-90. The patches, tailored to the tumor's shape and depth, aim to minimize damage to healthy tissues and structures. Through 3D printing technology we demonstrated the feasibility of creating individualized applicators filled with radioisotopes, that have the potential to deliver therapeutic doses while ensuring patient comfort and affordability. This innovative approach, based on radiation delivery simulations, offers a promising future alternative to conventional radiation therapy methods. Clinical trials are required to validate its efficacy and safety, but the potential for reduced radiation exposure and enhanced treatment precision positions this personalized and targeted therapy as a valuable advancement in skin cancer treatment.

Human-to-human communication via the computer is mainly done using a keyboard or microphone. In the field of Virtual Reality (VR), where the most immersive experience possible is desired, the use of a keyboard contradicts this goal, while the use of a microphone is not always desirable (e.g. silent commands during task force training) or simply not possible (e.g. if the user has a hearing loss). Data gloves help to increase immersion within the VR as they correspond to our natural interaction. At the same time, they offer the possibility to accurately capture hand shapes, such as those used in non-verbal communication (e.g. thumbs up, okay gesture, ...) and in sign language. In this paper, we present a hand shape recognition system using Manus Prime X data gloves, including data acquisition, data preprocessing, and data classification to enable nonverbal communication within VR. We investigate the impact on accuracy and classification time of using an Outlier Detection and a Feature Selection approach in our data preprocessing. To obtain a more generalized approach, we also studied the impact of artificial Data Augmentation, i.e., we create new artificial data from the recorded and filtered data to augment the training dataset. With our approach, 56 different hand shapes could be distinguished with an accuracy of up to 93.28%. With a reduced number of 27 hand shapes, an accuracy of up to 95.55% could be achieved. Voting Meta-Classifier (VL2) has proven to be the most accurate, albeit slowest, classifier. A good alternative is Random Forest (RF), which was even able to achieve better accuracy values in a few cases and was generally somewhat faster. Outlier Detection has proven to be a effective approach, especially in improving classification time. Overall, we have shown that our hand shape recognition system using data gloves is suitable for communication within VR.

This comprehensive study focused on the standard taxi sign used in Ireland, its impact on drag production, fuel expenses, and CO2 emissions. Experimental analysis revealed that the conventional taxi sign significantly increased drag, especially when mounted on streamlined vehicles such as saloon cars, due to flow separation issues on the rear roof and rear windshield. Longitudinal reorientation of the sign offered a 14-fold reduction in drag increase compared to the traditional placement. It was found that positioning the sign in the middle of the roof offered the greatest fuel efficiency. Furthermore, the study estimated that implementing longitudinal repositioning on all Irish taxi signs could save drivers approximately 832 per year and reduce national CO2 emissions by a substantial 22,464 tonnes annually. Comparative analyses with international taxi signs demonstrated that the Irish sign had significantly larger drag contributions, emphasizing the need for improved aerodynamics. To address the inherent drag issue, the study explored novel appendable devices and proposed alternative taxi sign designs. Among the tested solutions, a magnet-mounted front ramp proved the most effective, reducing total drag by nearly 30%. Additionally, a motorized flip-up taxi sign design demonstrated a remarkable 40% reduction in drag. Finally, a newly proposed taxi sign design, featuring longitudinal positioning and pointed triangular front and rear faces, resulted in a minimal 4.3% increase in vehicle drag compared to the baseline car.

American Sign Language is a popular language for deaf individuals. Communication is made easier for these people through sign language. However, in a digital era like today, there is a need for these people to be able to communicate online, and even get help from technology to communicate in person with non sign language speakers. This research will present a program able to translate American sign language to plain English. This study aims to use the OpenCV library to recognize hand signals, also a trained model to identify images so that the program can then translate them to words and letters. The program uses a data set of over 2000 images which will be in this case the largest data set available. With over 90\% of accuracy it results in a basic computer program with the largest data set available that would make possible for users to communicate with a wide variety of words and expressions.

The move from paper to online is not only necessary for remote working, it is also significantly more sustainable. This trend has seen a rising need for high-quality digitization of content from pages and whiteboards to sharable online material. But capturing this information is not always easy, nor are the results always satisfactory. Available scanning apps vary in their usability and do not always produce clean results, retaining surface imperfections from the page or whiteboard in their output images. CleanPage, a novel smartphone-based document and whiteboard scanning system, is presented. CleanPage requires one button-tap to capture, identify, crop and clean an image of a page or whiteboard. Unlike equivalent systems, no user intervention is required during processing and the result is a high-contrast, low-noise image with a clean homogenous background. Results are presented for a selection of scenarios showing the versatility of the design. CleanPage is compared with two market leader scanning apps using two testing approaches: real paper scans and ground-truth comparisons. These comparisons are achieved by a new testing methodology that allows scans to be compared to unscanned counterparts, by using synthesized images. Real paper scans are tested using image quality measures. An evaluation of standard image quality assessments is included in this work and a novel quality measure for scanned images is proposed and validated. The user experience for each scanning app is assessed, showing CleanPage to be fast and easier to use.

More than 700 anthocyanins are described so far, and many different sources are rich in these natural pigments. During the last few years, researchers have been focusing of strategies to recover and purify anthocyanins, from simple liquid-liquid extractions to more advanced supercritical CO2 extractions, however, with such a huge number of different sources, it’s extremely hard to get a simple and fast methodology that fits several sources, not only by the structural nature of the anthocyanins but also by the presence of specific groups of natural compounds. Using simple methodologies like solid-phase extraction resins represents a powerful economical and efficiency strategy. In this work, we tested several sources of structurally distinct anthocyanins, including purple basil, purple cabbage, purple sweet potato, butterfly pea flower, cornflower and wild pansy, in different resins to assess their efficiency in the obtention of anthocyanin-pure extracts. The results showed us that the combination of oasis HLB® resin followed by a cation-exchange SPE with DSC-MCAX® resulted in high-purity anthocyanin extracts for all the cases.

We propose a neural-network-based watermarking method that introduces the quantized activation function that approximates the quantization of JPEG compression. Many neural-network-based watermarking methods have been proposed. Conventional methods have acquired robustness against various attacks by introducing an attack simulation layer between the embedding network and the extraction network. The quantization process of JPEG compression was replaced by the noise addition process in the attack layer of the conventional methods. In this paper, we propose a quantized activation function that can simulate the JPEG quantization standard as it is in order to improve the robustness against the JPEG compression. Our quantized activation function consists of several hyperbolic tangent functions and is applied as an activation function for neural networks. Our network was introduced in the attack layer of ReDMark proposed by Ahmadi et al. to compare it with their method. That is, the embedding and extraction networks had the same structure. We compared the usual JPEG compressed images and the images applying the quantized activation function. The results showed that a network with quantized activation functions can approximate JPEG compression with high accuracy. We also compared the bit error rate (BER) of estimated watermarks generated by our network with those generated by ReDMark. We found that our network was able to produce estimated watermarks with lower BERs than those of ReDMark. Therefore, our network outperformed the conventional method with respect to image quality and BER.

The branch prediction units (BPUs) generally have security vulnerabilities, which can be used by attackers to obtain the execution status, jump directions, and jump address of conditional branches, and the existing protection methods cannot defend against these attacks. Therefore, this article proposes a hardware security protection method for conditional branches of embedded systems. This method calculates the number of updates to the branch target buffer (BTB). When it exceeds the threshold, BTB is locked to prevent attackers from analyzing the execution status of branches based on the time difference of whether BTB is updated. Moreover, the hybrid physical unclonable function (PUF) circuit is designed to provide confidentiality protection for the jump directions, jump addresses, and their indexes, preventing attackers from stealing these critical data. If these mechanism fails and attackers successfully tamper with conditional branches, this paper proposes a control flow integrity (CFI) protection mechanism based on branch labels to timely detect tampering with instruction codes, jump addresses, and jump directions. The proposed method is implemented and tested on FPGA. The experimental results show that this method can achieve fine-grained security protection for conditional branches, with about 5.4% resource overhead and less than 5.5% performance overhead.

Sign Language Recognition (SLR) aims to bridge speech-impaired and general communities by recognizing signs from given videos. Researchers still face challenges developing efficient SLR systems because of the video’s complex background, light illumination, and subject structures. Recently many researchers developed a skeleton-based sign language recognition system to overcome the subject and background variation of hand gesture signs. However, skeleton-based SLR is still under exploration due to the lack of information and annotations on hand key points. More recently, researchers included body and face information with the hand gesture for the SLR, but their performance and efficiency are unsatisfactory. We proposed a Multi-Stream Graph-based Deep Neural Network (SL-GDN) for a skeleton-based SLR system to overcome the problems. The main purpose of the proposed SL-GDN approach is to improve the efficiency and performance of the SLR system with a low computational cost based on the human body pose in the form of 2D landmark locations. In the procedure, firstly, we constructed a skeleton graph based on the selected 27 whole-body key points among 67 key points to solve the inefficiency problems. Then we proposed multi-stream SL-GDN to extract features from the whole-body skeleton graph for four streams. Finally, we concatenated the four different features and applied a classification module to refine the feature and recognize corresponding sign classes. Our data-driven and graph construction method increases the system’s flexibility and brings high generability to adapt various data samples. We used three large-scale benchmark SLR datasets to evaluate the proposed model: WLASL, AUTSL and CSL. The demonstrated performance accuracy table proved the superiority of the proposed model, and we believe this will be considered a great invention in the SLR domain.

This paper presents technical solutions to increase Industry 4.0 maturity. Within the “5G-Enabled Manufacturing” project, a 5G network has been deployed at the shop floor to enable fast and scalable connectivity. This network was used to connect a grinding machine to a private cloud to enable visibility and transparency of the production data, which is the basis for Industry 4.0 and smart manufacturing. Results indicate a present lack of commercially available product independent solutions for interconnection and data transfer despite the availability of open standards and well-documented demonstrator projects. The solution is described and discussed regarding technical interoperability, focusing on the system layout, communication standards, and open systems. From the discussion, it is derived that manufacturing end-users are in need to expand and further internalize knowledge on future information and communication technologies to reduce their dependency on equipment and technology providers.

The goal of background reconstruction is to recover the background image of a scene from a sequence of frames showing this scene cluttered by various moving objects. This task is fundamental in image analysis, and is generally the first step before more advanced processing, but difficult because there is no formal definition of what should be considered as background or foreground and the results may be severely impacted by various challenges such as illumination changes, intermittent object motions, highly cluttered scenes, etc. We propose in this paper a new iterative algorithm for background reconstruction, where the current estimate of the background is used to guess which image pixels are background pixels and a new background estimation is performed using those pixels only. We then show that the proposed algorithm, which uses stochastic gradient descent for improved regularization, is more accurate than the state of the art on the challenging SBMnet dataset, especially for short videos with low frame rates, and is also fast, reaching an average of 52 fps on this dataset when parameterized for maximal accuracy using GPU acceleration and a Python implementation.

Bayesian formulations of learning imply that whenever the evidence for a correlation between events in an animal’s habitat is sufficient, the correlation is learned. This implies that regularities can be learnt rapidly, from small numbers of learning examples. This speed of learning gives maximum possible fitness, and no faster learning is possible. There is evidence in many domains that animals and people can learn at nearly Bayesian optimal speeds. These domains include associative conditioning, and the more complex domains of navigation and language. There are computational models of learning which learn at near-Bayesian speeds in complex domains, and which can scale well – to learn thousands of pieces of knowledge (i.e., relations and associations). These are not neural net models. They can be defined in computational terms, as algorithms and data structures at David Marr’s [1] Level Two. Their key data structures are composite feature structures, which are graphs of multiple linked nodes. This leads to the hypothesis that animal learning results not from deep neural nets (which typically require thousands of training exam-ples), but from neural implementations of the Level Two models of fast learning; and that neu-rons provide the facilities needed to implement those models at Marr’s Level Three. The required facilities include feature structures, dynamic binding, one-shot memory for many feature struc-tures, pattern-based associative retrieval, unification and generalization of feature structures. These may be supported by multiplexing of data and metadata in the same neural fibres.

The segmentation of abnormal regions is vital in smart manufacturing. However, the existing segmentation system for detecting sauce-packet leakage on intelligent sensors encounters an issue of imaging blurring caused by uneven illumination. This issue adversely affects segmentation performance, thereby impeding the rapid production of industrial assembly lines. To alleviate this issue, we propose the two-stage Illumination-aware Sauce-packet Leakage Segmentation (ISLS) method for intelligent sensors. The ISLS comprises two main stages: Illumination-aware region enhancement and leakage region segmentation. In the first stage, YOLO-Fastestv2 is employed to capture the Region of Interest (ROI), which reduces redundancy computations. Additionally, we propose an image enhancement to relieve the impact of uneven illumination, enhancing the texture details of ROI. In the second stage, we propose a novel feature extraction network. Specifically, we propose the Multi-scale Feature Fusion Module (MFFM) and the Sequential Self-Attention Mechanism (SSAM) to capture discriminative representations of leakage. The multi-level features are fused by MFFM with a small number of parameters, which capture leakage semantics at different scales. The SSAM realizes the enhancement of valid features and the suppression of invalid features by adaptive weighting of spatial and channel dimensions. Furthermore, we generated a self-built dataset of sauce-packets, including 606 images with various leakage areas. Comprehensive experiments demonstrate that our ISLS method shows better results than several state-of-the-art methods, with additional performance analyses deployed on intelligent sensors to affirm the effectiveness of our proposed method. Our code is available at https://github.com/LSJ5106/SauceDetect.

Brand trust holds significant importance in the Mongolian cashmere industry. This study examines the primary predictors of customer brand trust among Mongolian cashmere users. Our research includes a survey with 800 participants and 20 individual interviews conducted among both Mongolian cashmere users and non-users. We utilized an online platform to reach participants, employing convenience sampling. The study investigates the relationship between various factors and brand trust within the context of Mongolian cashmere brands. Through a combination of in-dividual interviews and statistical analysis, we test six hypotheses. The findings indicate that higher brand quality, a strong reputation, greater transparency, effective issue resolution, commitment to environmental and social responsibility, and pre-purchase research on such commitments positively impact brand trust. Key insights gleaned from the interviews emphasize the importance of quality, reputation, transparency, customer service, and environmental responsibility in shaping consumer trust. These results offer valuable implications for cashmere brands aiming to bolster consumer trust and foster loyalty.

Strip steel surface defect recognition research has important research significance in industrial production. Aiming at the problems of defect feature extraction difficulty, slow detection speed and insufficient dataset, YOLOv5 is improved on the basis of YOLOv5, and YOLO-LFPD (Lightweight fine particle detection) model is proposed. By introducing the RepVGG (Re-param VGG) module, the robustness of the model to the surface defect features of strip steel is enhanced, and the expressive ability of the model is improved. FasterNet is used to replace the backbone network, which ensures the accuracy and accelerates the inference speed, making the model more suitable for real-time monitoring. The use of pruning, GA genetic algorithm with OTA loss function method further reduces the model size while better learning the strip steel defect feature information, thus improving the generalisation ability and accuracy of the model. The experimental results show that the introduction of the RepVGG module and the use of FasterNet can well improve the model performance, with a reduction of 48% in the number of parameters, a reduction of 13% in the number of GFLOPs, an inference time of 77% of the original, and an optimal accuracy compared with the network models in recent years. The experimental results on the NEU-DET dataset show that the accuracy of YOLO-LFPD is improved by 3% to 81.2%, which is better than other models, and provides new ideas and references for the lightweight strip steel surface defect detection scenarios and application deployment.

In emergency situations, ensuring standardized Cardiopulmonary Resuscitation (CPR) actions is crucial. However, current Automated External Defibrillators (AEDs) lack methods to determine whether CPR actions are performed correctly, leading to inconsistent CPR quality. To address this issue, we introduce a novel method called Deep Learning-based CPR Action Standardization (DLCAS). This method involves three parts. First, it detects correct posture using OpenPose to recognize skeletal points. Second, it identifies a marker wristband with our CPR-Detection algorithm and measures compression depth, count and frequency using a depth algorithm. Finally, we optimize the algorithm for edge devices to enhance real-time processing speed. Extensive experiments on our custom dataset have shown that the CPR-Detection algorithm achieves a mAP0.5 of 97.04%, while reducing parameters to 0.20M and FLOPs to 132.15K. In a complete CPR operation procedure, the depth measurement solution achieves an accuracy of 90% with a margin of error less than 1 cm, while the count and frequency measurements achieve 98% accuracy with a margin of error less than 2 counts. Our method meets the real-time requirements in medical scenarios, and the processing speed on edge devices have increased from 8fps to 25fps.

This study considers ways to increase the consumption of sustainable fashion, given the significant environmental and social damages, associated with the industry. A series of experiments were conducted examining the impacts of choice architecture (nudges) under field conditions in collab-oration with one of Israel’s largest shopping centers. The study sought to identify which inter-ventions at the retail level successfully motivate sustainable fashion behavioral change regarding purchases and willingness to pay more, along with agreement with several statements regarding the climate crisis and sustainable fashion. Among the types of nudges examined in this field study were providing information, increasing accessibility to sustainable alternatives and appealing to social identity, in relation to demographics and green self-image. The study found that offering alterna-tives to consumers constitutes the most effective way to “nudge” consumers toward more sus-tainable purchasing behavior. Nonetheless, this does not negate the contribution of providing in-formation and strengthening social norms regarding sustainable fashion. Additionally in all groups, most participants reported that they do not know how to distinguish between sustainable and non-sustainable fashion, nor do they believe that the clothes they purchased were actually sustainable. The findings emphasize the need for policies that will increase the accessibility of sustainable fashion.

Applications of gradient method for nonlinear optimization in development of Gradient Neural Network (GNN) and Zhang Neural Network (ZNN) are investigated. Particularly, the solution of the matrix equation AXB=D which changes over time is studied using the novel GNN model, termed as GGNN(A,B,D). The GGNN model is developed applying GNN dynamics on the gradient of the error matrix used in the development of the GNN model. The convergence analysis shows that the neural state matrix of the GGNN(A,B,D) design converges asymptotically to the solution of the matrix equation AXB=D, for any initial state matrix. It is also shown that the convergence result is the least square solution which is defined depending on the selected initial matrix. A hybridization of GGNN with analogous modification GZNN of the ZNN dynamics is considered. The Simulink implementation of presented GGNN models is carried out on the set of real matrices.

Initial choice of Learning Rate is a key part of gradient based methods and has a great effect on the performance of the Deep Learning Model.This paper studies the behavior of multiple gradient based optimization algorithm which are commonly used in Deep Learning and compare their performance on various learning rate. As observed popular choice of optimization algorithms are highly sensitive to various choice of learning rates. Our goal is to find which optimizer has an edge over others for a specific setting. We look at two datasets namely MNIST and CIFAR10 for benchmarking. The results are quite surprising, and it will help us to choose a learning rate more efficiently.

Daltonisation refers to the recolouring of images such that details normally lost by colour vision deficient observers become visible. This comes at the cost of introducing artificial colours. In a previous work, we presented a gradient-domain colour image daltonisation method that outperformed previously known methods both in behavioural and psychometric experiments. In the present paper, we improve the method by (i) finding a good first estimate of the daltonised image, thus reducing the computational time significantly, and (ii) introducing local linear anisotropic diffusion, thus effectively removing the halo artefacts. The method uses a colour vision deficiency simulation algorithm as an ingredient, and can thus be applied for any colour vision deficiency, and can even be individualised if the exact individual colour vision is known.

Artificial Intelligence is dominated by Artificial Neural Networks (ANNs). Currently, the Batch Gradient Descent (BGD) is the only solution to train ANN weights when dealing with large datasets. In this article, a modification to the BGD is proposed which significantly reduces the training time and improves the convergence. The modification, called Instance Eliminating Back Propagation (IEBP), eliminates correctly-predicted-instances from the Back Propagation. The speedup is due to the elimination of unnecessary matrix multiplication operations from the Back Propagation. The proposed modification does not add any training hyperparameter to the existing ones and reduces the memory consumption during the training.

Video quality assessment (VQA) plays an important role in video applications for quality evaluation and resource allocation. It aims to evaluate the video quality consistent with the human perception. In this letter, a hierarchical gradient similarity based VQA metric is proposed inspired by the structure of the primate visual cortex, in which visual information is processed through sequential visual areas. These areas are modeled with the corresponding measures to evaluate the overall perceptual quality. Experimental results on the LIVE database show that the proposed VQA metric significantly outperforms the state-of-the-art VQA metrics.

Background: Machine learning can analyze vast amounts of data and make predictions for events in the future. Our group created machine learning models for vital sign predictions. To transport the information of these predictions without numbers and numerical values and make them easily usable for human caregivers, we aimed to integrate them into the Philips Visual-Patient-avatar, an avatar-based visualization of patient monitoring. Methods: We conducted a computer-based simulation study with 70 participants in three European university hospitals. We validated the vital sign prediction visualizations by testing their identification by anesthesiologists and intensivists. Each prediction visualization consisted of a condition (e.g., blood pressure low) and an urgency (a visual indication of the timespan in which the condition is expected to occur). To obtain qualitative user feedback, we also conducted standardised interviews and derived statements that participants later rated in an online survey. Results: The mixed logistic regression model showed 77.9% (95%CI 73.2-82.0%) correct identification of prediction visualizations (i.e. condition and urgency both correctly identified) and 93.8% (95%CI 93.7-93.8%) for conditions only (i.e. without considering urgencies). Forty-nine of 70 participants completed the online survey. The online survey participants agreed that the prediction visualizations were fun to use (32/49, 65.3%), and that they could imagine working with them in the future (30/49, 61.2%). They also agreed that identifying the urgencies was difficult (32/49, 65.3%). Conclusions: This study found that care providers correctly identified >90% of the conditions (i.e. without considering urgencies). The accuracy of identification decreased when considering urgencies in addition to conditions. Therefore, in future development of the technology, we will focus on either only displaying conditions (without urgencies) or improve the visualizations of urgency to enhance usability for human users.

Postoperative adhesive arachnoiditis is an inflammatory response of the spinal leptomeninges that occurs after surgery, resulting in scar formation in the avascular nature of the arachnoid layer. Clinical manifestations of this condition include pain, sensory deficits, motor dysfunction, reflex abnormalities, and bladder/bowel impairments. In MRI scans, the signs of postoperative adhesive arachnoiditis can vary, but some specific indicators can assist surgeons in locating the lesion and planning surgical interventions more accurately. In this report, we reported a 37-year-old man with postoperative adhesive arachnoiditis after two surgeries for Chiari I malformation, demonstrating that the gradually-forming the delta cord sign with sequential syringomyelia. During the intraoperative examination, we identified the presence of the delta cord sign, which was formed by an arachnoid scar that was tethering the dorsal spinal cord to the dura. This discovery allowed us to precisely pinpoint the location of the arachnoid scar, thereby providing valuable guidance to avoid any unnecessary exploration of innocent structures during the procedure. Other localization signs were also reviewed.

Abstract In order to solve the problem that it is difficult to obtain the accurate trigger signal moment when using the time difference of UAV flight acoustic signals collected by acoustic sensors arranged in advance to locate the position of UAVs when they cannot be tracked by radar and are difficult to be observed by the human eye, an acoustic source localisation method with improved empirical modal decomposition (REMD) under an adaptive frequency window is proposed. Firstly, the collected UAV flight signals are smoothed and filtered, then the robust empirical modal decomposition (REMD) is used for IMF modal decomposition and spectral analysis of the collected signals, and a panning frequency window with variable bandwidth is introduced for corresponding frequency locking and extracting the IMFs; the GWO optimisation is used and the sliding metrics are set to determine the position of the panning frequency window automatically; finally, the extracted Finally, the extracted IMF components are reconstructed according to the criterion, and the trigger signal moments in the reconstructed IMF components are resolved and recorded, and the algorithm is used to calculate the difference between the reception moments of different sensors; for the traditional sound source localisation algorithms with a small detection range, which are prone to fall into the no-solution or false-solution, i.e., unable to achieve the localisation, a weighted least-squares-based Chan-Taylor algorithm is proposed for the localisation computation, and a weighted least-squares algorithm is proposed for the localisation computation, through the Inputting the sensor delay parameter and solving the linear equation system to get the target position information; finally, this paper verifies the robustness and performance of the design method with simulated and actual signals. The results show that the maximum positioning error is no more than 5% in an area with a side length of 15m in the measurement area. With the improved accuracy of time delay estimation, the positioning simulation results meet the positioning requirements even though the positioning error is further expanded when the measurement area is further expanded.

Rapid and accurate detection of road cracks is of great significance for road health monitoring, but currently this work is mainly completed through manual site surveys. Low-altitude UAV remote sensing can obtain images with centimeter or even subcentimeter ground resolution, which provides a new efficient and economical approach for rapid crack detection. Nevertheless, crack detection networks face challenges such as edge blurring and misidentification due to the heterogeneity of road cracks and the complexity of the background. To address these issues, we propose a real-time edge reconstruction crack detection network (ERNet) which adopts multi-level information aggregation to reconstruct crack edges and improve the accuracy of segmentation between the target and background. To capture global dependencies across spatial and channel levels, we propose an efficient bilateral decomposed convolutional attention module (BDAM) that combines depth separable convolution and dilated convolution to capture global dependencies across spatial and channel levels. To enhance the accuracy of crack detection, we use a coordinate-based fusion module that integrates spatial, semantic, and edge reconstruction information. In addition, we propose an automatic measurement of crack information for extracting the crack trunk and its corresponding length and width. Experimental results demonstrate that our network achieves the best balance between accuracy and inference speed.

Rhombencephalosynapsis (RES) is a rare, congenital developmental disorder of the central nervous system. Even after many years of its identification, the etiology and pathogenesis of RES remain unclear. It occurs in combination with many syndromes and thus has overlapping symptoms. The clinical presentation of this disorder varies depending on the severity of the maldevelopment of vermis. Here, we present a case of a 60-year-old Indian female diagnosed with Rhombencephalosynapsis. Clinical and MRI characteristics of RES along with treatment options have been discussed.

A particular challenge in the conceptual design of direct-from orbit delivery systems is the seemingly well-known kinematics retain fallacies whose efficacies are reduced over great distances. Rotation about the local wing of an aerospace vehicle is almost never the pitch angle, yet modern application of kinematics often assumes such (with accompanying angular error). The same assertion is usually true about the nature of roll and yaw angles. Expressing motion in coordinates of rotating reference frames necessitates transformation between reference frames, and one such transformation is embodied in the Direction Cosine Matrices (DCM) formed by a sequence of three successive frame rotations. One of two ubiquitous sequences of three rotations used to construct the DCM involves first a rotation around the inertial z-axis, then intermediate y-axis, then finally about the body’s x-axis; a sequence commonly called the “aerospace sequence” or the a “3-2-1 rotation sequence”. The second ubiquitous sequence is the so-called “orbital sequence” or “3-1-3 rotation sequence” with rotations about the inertial z axis first, then about the intermediate x-axis, then finally about the body z-axis. This manuscript evaluates which sequence is the most advantageous for an object starting in space and then travels through the atmosphere to a target on the Earth’s surface. Six degrees of freedom of vehicle motion were simulated starting in orbit with a given thrust and commanded maneuver. The simulation performs all twelve possible rotation sequences (transforming inertial coordinates to body coordinates) with comparison by computational burden and error representing rotations about body axes (roll, pitch, and yaw respectively). Simulation precision is validated using the quaternion normalization condition indicating near machine precision (0.9\times{10}^{-15}) and reveals the so-called 132 rotation is the most accurate with an average error of 0.14° and a computational time of 0.013 seconds: resulting in a 97.95% increase in accuracy over the so-called 321 rotation and a 99.84% increase over the so-called 313 rotation.

Model understanding is critical in many domains, particularly those involved in high-stakes decisions, i.e., medicine, criminal justice, and autonomous driving. Explainable AI (XAI) methods are essential for working with black-box models such as Convolutional Neural Networks. This paper evaluates the traffic sign classifier of Deep Neural Network (DNN) from the Programmable Systems for Intelligence in Automobiles (PRYSTINE) project for explainability. The results of explanations were further used for the CNN PRYSTINE classifier vague kernels` compression. After all, the precision of the classifier was evaluated in different pruning scenarios. The proposed classifier performance methodology was realised by creating the original traffic sign and traffic light classification and explanation code. First, the status of the kernels of the network was evaluated for explainability. For this task, the post-hoc, local, meaningful perturbation-based forward explainable method was integrated into the model to evaluate each kernel status of the network. This method enabled distinguishing high and low-impact kernels in the CNN. Second, the vague kernels of the classifier of the last layer before the fully connected layer were excluded by withdrawing them from the network. Third, the network's precision was evaluated in different kernel compression levels. It is shown that by using the XAI approach for network kernel compression, the pruning of 5% of kernels leads only to a 1% loss in traffic sign and traffic light classification precision. The proposed methodology is crucial where execution time and processing capacity prevail.

The monocular structured light measurement system has low cost, simple algorithms, and a wide measurement range, making it widely applicable in multiple fields. However, in practical applications, the system may be affected by various factors including noise, non-linear intensity, changes in the reflectivity of the object being measured, and the calibration method and accuracy of the system, thereby reducing the measurement accuracy. In order to solve these issues, this paper proposes a new monocular structured light measurement method, mainly comprises the following two points: (a) A denoising algorithm for phase images based on DAE is proposed, which calculates the wrapped phase of the phase-shifted fringes after initial denoising by DAE. Then, a new phase-shifted fringe pattern is generated and input into DAE for iterative denoising, achieving high-performance image denoising and high-precision wrapped phase solution. (b) A new absolute phase height calibration algorithm is proposed, which introduces camera internal and external parameters and uses a two-layer feedforward network to directly establish the relationship between phase and the 3D coordinate system. Without the need for high-precision motion platforms, high-precision phase height calibration is attainable. In general, compared to conventional methods, the experimental results indicate the effectiveness of the proposed method for low-quality phase-shifted fringe 3D reconstruction. In addition, the measurement method described in this article also demonstrates effectiveness in high-dynamic scenes.

As argued by ‘Declaration of Concrete Environment (2010)’ of Korea and ‘Declaration of Asian Concrete Environment (2011)’ of six Asian countries, concrete as a single material has lately shown extremely large impact on environmental issues such as climate change. Assessment of environmental impact from concrete material and production has considerable importance. Concrete is a major material used in the construction industry that emits a large amount of substances with environmental impacts during its life cycle. Accordingly, technologies for the reduction in and assessment of the environmental impact of concrete from the perspective of Life Cycle Assessment must be developed. At present, the studies in relation to greenhouse gas emission from concrete are being carried out globally as a countermeasure against climate change. In this study, a sustainable concrete mix design algorithm was designed using correlation analyses, and its carbon emission and cost reduction performances were assessed. Using correlation analyses, the concrete strength, w/b and s/a ratios, and CO₂ emissions were identified as major variables of concrete mix design that influenced other variables. Also, this study aims to evaluate the CO₂ emission reduction performance of the algorithm-deduced sustainable concrete mix design, and therefore, the CO₂ emissions of the sustainable concrete mix design are compared with those of the actual concrete mix design applied to the construction of the office building A in South Korea.

Fast neutron (FN) radiation mediated mutagenesis is a unique approach among the several induced mutagenesis methods being used in plant science in terms of impacted mutations. The FN mutagenesis usually creates deletions from a few bases to several million bases (Mb). A library of random deletion generated using FN mutagenesis lines can provide indispensable resources for the reverse genetic approaches. In this review, information from several efforts made using FN mutagenesis has been compiled to understand the type of induced mutations, frequency, and genetic stability. Concerns regarding the utilization of FN mutagenesis technique for a plant with different level of ploidy and genome complexity are discussed. We have highlighted the utility of next-generation sequencing techniques which can be efficiently utilized for the characterization of mutant lines as well as for the mapping of causal mutations. Pros and cons of Mut-map, MutChromSeq, exon capture, whole genome sequencing, MutRen-Seq, and different tilling approaches can be used for the detection of FN-induced mutation has also been discussed. Genomic resources developed using the FN mutagenesis have been catalogued wooing to meaningful utilization of the available resources. The information provided here will be helpful for the efficient exploration for the crop improvement programs and for better understanding of genetic regulations.

Fast Field-Cycling (FFC) Nuclear Magnetic Resonance (NMR) relaxometry is a powerful non-destructive magnetic resonance technique that enables, among other things, the investigation of slow molecular dynamics at low magnetic field intensities. FFC-NMR relaxometry measurements provide insight into molecular motion across various timescales within a single experiment. This study focuses on a \textit{model-free} approach, representing the NMRD profile $R_1$ as a linear combination of Lorentzian functions, thereby addressing the challenges of fitting data within an ill-conditioned linear least-squares framework. Tackling this problem, we present a comprehensive review and experimental validation of three regularization approaches to implement the model-free approach to analysing NMRD profiles. These include: (1) \nameUpe, utilizing locally adapted $L_2$ regularization; (2) \nameL, based on $L_1$ penalties; and (3) a hybrid approach combining locally adapted $L_2$ and global $L_1$ penalties. Each method's regularization parameters are determined automatically, according to the Balancing and Uniform Penalty principles. % Our contributions include the implementation and experimental validation of the \nameUpe{} and \nameUp{} algorithms, and the development of a "dispersion analysis" technique to assess the existence range of the estimated parameters. The objective of this work is to delineate the variance in fit quality and correlation time distribution yielded by each algorithm, thus broadening the set of software tools for the analysis of sample structures in FFC-NMR studies. The findings underline the efficacy and applicability of these algorithms in the analysis of NMRD profiles from samples representing different potential scenarios. 1

The accompanying mechanical behavior during the flow of ore-rock dispersion is an important factor leading to the instability and failure of the orepass in the ore storage section. It is of great significance to accurately understand the instability characteristics of the well wall in the ore section, during the ore drawing process by understanding the flow characteristics and internal mechanical transfer mechanism of ore-rock dispersion. Using the discrete element method to analyze the flow characteristics, contact compactness, stress distribution characteristics and contact force probability distribution of the ore-rock dispersion, it realizes the quantitative characterization of the damage degree of ore-rock flow, and reveals the damage mechanism of mine storage wall. The results show that: (1) The friction between ore-rock particles, between ore-rock and well wall is an important reason for the change of the macroscopic flow form of ore-rock from “one” to “V” type. (2) The compactness and stress concentration of ores are increasing with the increase of storage depth, which is shown that with the increase of storage depth, the greater the probability of well wall instability. (3) The probability distribution of contact force strength between particles decreases exponentially in the whole process of ore drawing, in which the strong chain plays a major role in the stability of the dispersion system. (4) The number of overpressure and the overpressure coefficient can be used to quantitatively characterize the wall damage degree under the action of ore-rock flow. The dynamic side pressure decreases periodically in the form of exponential fluctuations, and the dynamic load formed by the ore-rock flow mainly played a role in the lower area of the ore storage section.

The wellness economy has developed rapidly recently, particularly in wellness tourism. Wellness tourism transforms sustainable strategy through the availability of extended multi-actor engagement in service innovation to further the tourism experience. Therefore, when operating a wellness tourism destination, it is necessary to encourage multi-actor engagement in service innovation to achieve sustainable development. This paper introduces a mixed research method with the SWOT-AHP-QSPM, as it qualitatively analyzes the advantages, disadvantages, opportunities, and threats, constructs an evaluation index system, conducts quantitative analysis using the analytic hierarchy process, determines the index weight and weighted score, and calculate the sustainable development strategy, which is appropriate for wellness development of Chongqing Tea Mountain and Bamboo Forest National Forest Park. Purposive field surveys and personal interviews were conducted in Chongqing forest tourism destinations, involving 356 effective respondents. The research reveals that multi-actor engagement significantly enhances sustainable development. These findings bridge service innovation knowledge gaps and provide valuable insights for forest tourism destination management and sustainable policy strategies to elevate multi-actor engagement in forest sites.

Tack is one of the most important factors affecting the operability of prepregs, but it is still very difficult to obtain the change of prepreg tack with time at different temperatures. In this paper, a newly designed T-peel device is used to test the tack decay rate of prepregs at different temperatures. The obtained decay rate deviates less than 8.9% from the manufacturer's data. The maximum deviation of the ratio of tack decay speeds calculated by the Arrhenius equation from the experiments is not higher than -9.7%. It is proved that the present method can obtain the decay rate of tack of prepregs at different temperatures. This method is also applicable to the other two types of prepregs. The article proposes the handling life unit, an operational life measurement unit for prepregs, and establishes that the relationship between tack and remaining shelf life reduces prepreg management costs.

The development of computational acoustics allows simulation of sound generation and propagation in complex environment. In particular, meshfree methods are widely used to solve acoustics problems through arbitrarily distributed field points and approximation smoothness flexibility. As a Lagrangian meshfree method, smoothed particle hydrodynamics (SPH) method reduce the difficulty in solving problems with deformable boundaries, complex topologies, or multiphase medium. The traditional SPH method has been applied in acoustic simulation. This study presents the corrective smoothed particle method (CSPM), which is a combination of SPH kernel estimate and Taylor series expansion. The CSPM is introduced as a Lagrangian approach to improve accuracy in solving acoustic wave equations in the time domain. Moreover, a boundary treatment technique based on the hybrid meshfree and finite difference time domain (FDTD) method is proposed to represent different acoustic boundaries with particles. To model sound propagation in pipes with different boundaries, soft, rigid, and absorbing boundary conditions are built with this technique. Numerical results show that the CSPM algorithm is consistent and demonstrates convergence with exact solutions. Main computational parameters are discussed, and different boundary conditions are validated to be effective for benchmark problems in computational acoustics.

Glypicans are a family of heparan sulfate proteoglycans that are attached to the outer plasma membrane leaflet of the producing cell by a glycosylphosphatidylinositol anchor. Glypicans are involved in the regulation of many signalling pathways, including those that regulate the activities of Wnts, Hedgehog (Hh), Fibroblast Growth Factors (FGFs) and Bone Morphogenetic Proteins (BMPs), among others. In the Hh signalling pathway, glypicans have been shown to be essential for ligand transport and the formation of Hh gradients over long distances, for the maintenance of Hh levels in the extracellular matrix and for its unimpaired reception in distant recipient cells. Recently, two mechanistic models have been proposed to explain how Hh can form the signalling gradient and how glypicans may contribute to it. In this review, we describe the structure, biochemistry and metabolism of glypicans and their interactions with different components of the Hh signalling pathway that are important for the release, transport and reception of Hh.

The functioning of forest ecosystems depends on the taxonomic and ecological diversity of soil fungi. Urbanization is increasing worldwide and is regarded a major driver of environmental change altering local species assemblages in urban forest. We investigated whether the degree of urbanization and local forest characteristics affect the soil fungal community in 20 beech forests located along a rural–urban gradient in the city of Basel and its suburbs (Switzerland). We analyzed their soil fungal communities by DNA metabarcoding of the rDNA ITS2 region and related these data to local forest vegetation characteristics and soil properties. The number of fungal OTUs in the 20 forests examined ranged from 170 to 303. Richness, diversity and evenness of fungal OTUs were all significantly affected by the degree of urbanization, but in different ways. OTU richness was highest in forests in areas with a low degree of urbanization and lowest in forests in rural areas. In contrast, the diversity of OTUs increased with increasing degree of urbanization. Different fungal phyla and fungal guilds showed distinct patterns in their relative abundance along the rural-urban gradient. The degree of urbanization reduced the relative abundance of symbiotrophic fungi, but increased that of saprotrophic and pathotrophic fungi. Our results show that urbanization alters soil fungal community, which in turn can lead to changes in forest ecosystems.

Salinity acts as a critical environmental filter on microbial communities in natural systems, negatively affecting microbial diversity. However, how salinity affects the community assembly remains unclear. This study used Wendeng multi-pond saltern as a model to evaluate the prokaryotic community composition and diversity and quantify the relative importance of ecological processes across salinity gradients. Results showed that low saline salterns (45-80 g/L) exhibited higher bacterial diversity than those in high saline salterns (175-265 g/L). The relative abundance of taxa assigned to Halanaerobiaceae, Haloferacaceae, Desulfohalobiaceae, Phormidiaceae, Rohodobacteraceae, and Nitrococcaceae was higher with increasing salinity. Salinity and pH were the primary environmental factors that directly or indirectly determined the composition and diversity of prokaryotic communities. Microbial co-occurrence network dynamics were more complex in the sediment than in water of salterns. An infer Community Assembly Mechanisms by Phylogenetic-bin-based null model analysis (iCAMP) showed that microbial community assembly in sediment and water differed. Our findings provide more information about microbial community structure and the importance of various ecological processes in controlling microbial community diversity and succession along salinity gradients in water and sediment environments.

Hasselmann and coauthors proposed the discrete interaction approximation (DIA) as the best tool replacing the nonlinear evolution term in a numerical wind-wave model. Much later, Polnikov and Farina radically improved the original DIA by means of location all the interacting four wave vectors, used in the DIA configuration, exactly at the nodes of the numerical frequency-angular grid. This provides nearly two-times enhancing the speed of numerical calculation for the nonlinear evolution term in a wind-wave model. For this reason, the proposed version of the DIA was called as the fast DIA (FDIA). In this paper we demonstrate all details of the FDIA concept for several frequency-angular numerical grids of high resolution, with the aim of active implementation the FDIA in modern versions of world-wide used wind-wave models.