Wireless Sensor Networks (WSN) is designed to collect information across a large number of limited battery sensor nodes. Therefore, it is important to minimize the energy consumption of each node, which leads to the extension of the network life. Our goal is to design an intelligent WSN that collects as much information as possible to process it intelligently. To achieve this goal, an agent has been migrated to each node in order to process the information and to cooperate with these neighboring nodes while Mobile Agents (MA) can be used to reduce information between nodes and send those to the base station (Sink). This work proposes to use communication architecture for wireless sensor networks based on the Multi Agent System (MAS) to ensure optimal information collection. The collaboration of these agents generates a simple message that summarizes the important information in order to transmit it by a mobile agent. To reduce the size of the MA, the nodes of the network have been grouped into sector. As for each MA, we have established an optimal itinerary, consuming a minimum amount of energy with the data aggregation efficiency in a minimum time. Successive simulations in large scale wireless sensor networks through the SINALGO simulator show the performance of our proposal, in terms of energy consumption and package delivery rate.

Water covers a greater part of the earth's surface. Even though we know very little about the underwater world as most parts of it remain unexplored. Oceans including other water bodies hold huge natural resources and also the aquatic lives. These are mostly unexplored and very few of those are known due to unsuited and hazardous environments for the human to explore. This vast underwater world can be monitored remotely from a distant location with much ease and less risk. To monitor water-bodies remotely in real-time, sensor networking has been playing a great role. It is needed to deploy a wireless sensor network over the volume which we want to surveil. For vast water bodies like oceans, rivers and large lakes, data is collected from the different heights of the water level which is sent to the surface sink. Unlike terrestrial communication, radio waves and other conventional mediums can't serve the purpose of underwater communication as they pose high attenuation and very reduced transmission range. Rather an acoustic medium can transmit data more efficiently and reliably in comparison to other mediums. To transmit data reliably from the bottom of the sea to the sinks at the surface, multi-hop communication is needed which must involve a certain scheme. For seabed to surface sink communication, leading researchers have proposed different routing protocols. The goal of these routing protocols is to make underwater communication more reliable, energy-efficient and delay efficient thus to improve the performance of the overall communication. This paper surveys the advancement and applications of the routing protocols which eventually helps in finding the most efficient routing protocol for the Underwater Wireless Sensor Network (UWSN).

Underwater Wireless Sensor Networks (UWSN) enables various oceanic applications which require effective packet transmission. In this case, sparse node distribution, dynamic network topology and inappropriate selection of relay nodes cause void holes. Addressing this problem, we present a Relay based Void Hole Prevention and Repair protocol (ReVOHPR) by multiple Autonomous Underwater Vehicles (AUV) for UWSN. ReVOHPR efficiently identifies and avoids void holes and trap relay nodes to avoid it. ReVOHPR adopts the following operations as Ocean Depth (levels) based Equal Cluster Formation, Dynamic Sleep Scheduling, Virtual Graph based Routing, and Relay Assisted Void Hole Repair. For energy efficient cluster forming, Entropy based Eligibility Ranking (E2R) is presented which elects stable cluster heads (CHs). Then, dynamic sleep scheduling is implemented Dynamic Kernel Kalman Filter (DK2F) algorithm in which Sleep and Active modes based on the nodes current status. Inter Cluster Routing is performed by maximum matching nodes which selects by Dual criteria and also data transmitted to AUV. Finally, void holes are detected and repair by Bi-Criteria Mayfly Optimization (BiCMO) algorithm. The BiCMO focuses on reducing the number of holes, data packet loss and maximizes Quality of Service (QoS) and energy efficiency of the networks. This protocol is timely deal with node failures in packet transmission via multi-hop routing. Simulation is implemented by NS3 (AquaSim module) simulator that evaluates the performance in network simulation for following metrics as average energy consumption, delay, packet delivery rate and throughput.

Scaling the operation is the single greatest challenge in every enterprise. The more complex the enterprise, the greater the challenge. Scaling in healthcare has been a consistent challenge, with the need to meet clinical endpoints with less staff, increased costs of resources, regulatory oversight, and the goal of improving access to care. The article explores the emerging paradigm of multi-agent AI systems in healthcare, representing a significant leap beyond traditional Large Language Models. We comprehensively analyze the potential of these systems to revolutionize patient care, streamline administrative processes, and support complex clinical decision-making. The paper describes a hypothetical sepsis management system comprising seven specialized AI agents, each handling specific aspects of patient care, from data collection and diagnosis to treatment recommendations and resource management. We also examine applications in chronic disease management and hospital patient flow optimization. The technical implementation of these systems is discussed, including the use of advanced Large Language Models, inter-agent quality control measures, implementing guardrails, self-reflection, integration with Electronic Health Records, and the importance of explainable AI for decision transparency. In addition to describing the promising potential benefits, such as enhanced diagnostic accuracy and personalized treatment plans, we address significant challenges, including data quality assurance, workflow integration, and ethical considerations. We conclude the article by highlighting future directions, such as integrating IoT devices and developing more sophisticated natural language interfaces. Our work underscores the transformative potential of multi-agent AI systems in healthcare while emphasizing the need for rigorous validation, ethical oversight, and a patient-centered approach in their development and implementation.

Equality and incomparability multi-label ranking have not been introduced to learning before. This paper proposes new native ranker neural network to address the problem of multi-label ranking including incomparable preference orders using a new activation and error functions and new architecture. Preference Neural Network PNN solves the multi-label ranking problem, where labels may have indifference preference orders or subgroups which are equally ranked. PNN is a nondeep, multiple-value neuron, single middle layer and one or more output layers network. PNN uses a novel positive smooth staircase (PSS) or smooth staircase (SS) activation function and represents preference orders and Spearman ranking correlation as objective functions. It is introduced in two types, Type A is traditional NN architecture and Type B uses expanding architecture by introducing new type of hidden neuron has multiple activation function in middle layer and duplicated output layers to reinforce the ranking by increasing the number of weights. PNN accepts single data instance as inputs and output neurons represent the number of labels and output value represents the preference value. PNN is evaluated using a new preference mining data set that contains repeated label values which have not experimented on before. SS and PS speed-up the learning and PNN outperforms five previously proposed methods for strict label ranking in terms of accurate results with high computational efficiency.

Defect detection in ferromagnetic substrates is often hampered by non-magnetic coating thickness variation when using conventional eddy current testing technique. The lift-off distance between the sample and the sensor is one of the main obstacles for the thickness measurement of non-magnetic coatings on ferromagnetic substrates when using the eddy current testing technique. Based on the eddy current thin-skin effect and the lift-off insensitive inductance (LII), a simplified iterative algorithm is proposed for reducing the lift-off variation effect using a multi-frequency sensor. Compared to the previous techniques on compensating the lift-off error (e.g., the lift-off point of intersection) while retrieving the thickness, the simplified inductance algorithms avoid the computation burden of integration, which are used as embedded algorithms for the online retrieval of lift-offs via each frequency channel. The LII is determined by the dimension and geometry of the sensor, thus eliminating the need for empirical calibration. The method is validated by means of experimental measurements of the inductance of coatings with different materials and thicknesses on ferrous substrates (dual-phase alloy). The error of the calculated coating thickness has been controlled to within 3 % for an extended lift-off range of up to 10 mm.

This paper presents a new and enhanced fusion module for the Multi-Sensor Precipitation Estimator (MPE) that would objectively blend real-time satellite quantitative precipitation estimates (SQPE) with radar and gauge estimates. This module consists of a preprocessor that mitigates systematic bias in SQPE, and a two-way blending routine that statistically fuses adjusted SQPE with radar estimates. The preprocessor not only corrects systematic bias in SQPE, but also improves the spatial distribution of precipitation based on SQPE and makes it closely resemble that of radar-based observations. It uses a more sophisticated radar-satellite merging technique to blend preprocessed datasets, and provides a better overall QPE product. The performance of the new satellite-radar-gauge blending module is assessed using independent rain gauge data over a 5-year period between 2003-2007, and the assessment evaluates the accuracy of newly developed satellite-radar-gauge (SRG) blended products versus that of radar-gauge products (which represents MPE algorithm currently used in the NWS operations) over two regions: I) inside radar effective coverage and II) immediately outside radar coverage. The outcomes of the evaluation indicate a) ingest of SQPE over areas within effective radar coverage improve the quality of QPE by mitigating the errors in radar estimates in region I; and b) blending of radar, gauge, and satellite estimates over region II leads to reduction of errors relative to bias-corrected SQPE. In addition, the new module alleviates the discontinuities along the boundaries of radar effective coverage otherwise seen when SQPE is used directly to fill the areas outside of effective radar coverage.

In this paper, we present a multi-task model that predicts disparities and confidence levels in deep stereo matching simultaneously. We do this by combining its successful model for each separate task and obtaining a multitask model that can be trained with a proposed loss function. We show the advantages of this model compared to training and predicting disparity and confidence sequentially. This method enables an improvement of 15% to 30% in the area under the curve (AUC) metric when trained parallel rather than sequential. In addition, the effect of weighting the components in the loss function on the stereo and confidence performance is investigated. By improving the confidence estimate, the practicality of stereo estimators for creating distance images is increased.

This study explores fault-tolerant consensus in leader-following heterogeneous multi-agent systems, focusing on actuator failures in uncrewed aerial vehicles (UAVs) and uncrewed ground vehicles (UGVs). An optimization-based fault-tolerant consensus algorithm is proposed. The algorithm utilizes the Euler-Lagrange formula to ensure system consistency under actuator failures, with the Lyapunov stability theory proving the asymptotic stability of the consistency error. The algorithm is applied to heterogeneous multi-agent systems of UAVs and UGVs to derive optimal fault-tolerant consensus control laws for each vehicle type. Simulation experiments give evidence for the feasibility of the proposed control strategy.

Buildings are responsible for approximately 40% of the world’s energy consumption and 36% of the total carbon dioxide emissions. Building occupancy is essential, enabling Occupant-Centric Control for zero emissions and decarbonization. Although existing machine learning and deep learning methods for building occupancy prediction have achieved remarked progress, their analyses remain limited when applied to complex real-world scenarios. Besides, there is a high expectation for Transformer algorithms to predict building occupancy accurately. Therefore, this paper presents an Occupancy Prediction Transformer network (OPTnet). We fuse and feed multi-sensor data (building occupancy, indoor environmental conditions, HVAC operations) into a Transformer model to forecast the future occupancy presence in multiple zones. We perform experimental analysis and compare it to different occupancy prediction methods (e.g., Decision Tree, Long Short-Term Memory networks, Multi-layer perceptron) and diverse time horizons (1,2,3,5,10,20,30mins). The performance metrics (e.g., Accuracy and Mean Squared Error) are employed to evaluate the effectiveness of prediction algorithms. Our OPTnet method achieves superior performance on our experimental two-week data compared to existing methods. The improved performance signifies its potential to enhance HVAC control systems and energy optimization strategies. We will make the code publicly available at https://github.com/kailaisun/occupancy-prediction-binary to promote transparency and reproducibility.

Existing methods of hyperspectral anomaly detection still face several challenges: 1) Due to the limitations of self-supervision, avoiding the identity mapping of anomalies remains difficult; 2) The ineffective interaction between spatial and spectral features leads to the insufficient utilization of spatial information; 3) Current methods are not adaptable to the detection of multi-scale anomaly targets. To address the aforementioned challenges, we proposed a blind spot network based on multi-scale blind spot convolution for HAD. The multi-scale mask convolution module is employed to adapt to diverse scales of anomaly targets, while the dynamic fusion module is introduced to integrate the advantages of mask convolutions at different scales. The proposed approach includes a spatial-spectral joint module and a background feature attention mechanism to enhance the interaction between spatial-spectral features, with a specific emphasis on highlighting the significance of background features within the network. Furthermore, we propose a preprocessing technique that integrates pixel shuffle down-sampling (PD) operation with spatial spectral joint screening to address the issue of anomalous identity map-ping and facilitate finite-scale mask convolution to adapt to the detection of more scale targets. The proposed approach was assessed on four real hyperspectral datasets comprising anomaly targets of different scales. The experimental results demonstrate the effectiveness and superior performance of the proposed methodology compared with nine state-of-the art methods.

Along with the popularity of the Internet, people are exposed to more and more ways of micro-videos, and a huge amount of micro-video data has emerged. micro-videos have gradually become the Internet content preferred by the public, and a large number of micro-video apps have also emerged, such as Tiktok and Kwai. Intelligent classification and mining of micro-videos can greatly enhance user experience, improve business operation efficiency and enhance user experience. Through deep intelligent analysis and mining of micro-videos, important information in micro-videos can be extracted to provide an important basis for beautifying videos, content appreciation, video recommendation, content search, etc. In the past, content understanding for short videos often used human work annotation, but in recent years, with the great success of deep convolutional neural networks in image recognition, short video content understanding based on this method has gradually developed. Nowadays, most recognition algorithms extract the feature representation of each frame independently and then fuse them. However, while extracting the feature representation, some low-level semantic features are lost, which makes the algorithm unable to accurately distinguish the category of the video. At present, the algorithm of micro-video recognition based on deep learning has surpassed the iDT algorithm, making these traditional methods fade out of people’s view. In this paper according to the micro-video classification task, a new network model is proposed to concatenate features of each modality into the overall features of various modalities through the network, and then fuse the various modal features with the attention mechanism to obtain the whole micro-video features, which will be used for classification. In order to verify the effectiveness of the algorithm proposed in this paper, experiments are conducted in the public dataset, and it is shown the effectiveness of our model.

We propose a multi-sensor fusion localization framework for autonomous heavy-duty trucks suitable for mining scenarios, which enables high-precision, real-time trajectory generation and map construction. The motion estimated through pre-integration of the Inertial Measurement Unit (IMU) can eliminate distortions in the point cloud and provide an initial guess for LiDAR odometry optimization. The point cloud information obtained from LiDAR can assist in recovering depth information from image features extracted by the monocular camera. To ensure real-time performance, we introduce an iKD-Tree to organize the point cloud data. To address issues arising from bumpy road segments and long-distance driving in practical mining scenarios, we can incorporate a large number of relative and absolute measurements from different sources, such as GPS information and AprilTag-assisted localization data, including loop closure as factors in the system. The proposed method has been extensively evaluated on public datasets and self-collected datasets from mining sites.Our implementation is available at https://github.com/wangyicxy/iLM-slam

Distributed fiber optic vibration sensors based on Φ-OTDR using Rayleigh backscattering of highly coherent light source pulses are very popular type in the area of sensors due to many attractive features, mainly due to its distributed manner of sensing, i.e. they can operate as thousands of local sensors simultaneously. The paper summarizes and describes several important modifications of the original sensor scheme to enhance its key performance parameters and to extend its application potential. A multi-fiber sensor form was proposed. Single- and two-fiber sensors were designed, constructed and tested both in the laboratory and in a real environment. Variants of two-fiber sensor utilizing optical switches for the sensor flexibility enhancement are proposed and their applications are described.

Unmanned construction machinery vehicles mostly carry work in bridges, tunnels, and outdoor open spaces. Obtaining accurate pose estimation of the entire vehicle and establishing a map of the surrounding environment is of great significance for path planning and control in the later stage. Traditional simultaneous localization and mapping (SLAM) schemes, which mostly use a single sensor, but there are problems with localization drift and mapping failure in scenarios where there are few geometric features and the environment is prone to degradation. Currently, the multi-sensor fusion strategy has been proven to be an effective solution and widely used in the field of unmanned vehicle localization and mapping. This paper proposes a SLAM framework that tightly couples a LiDAR, IMU and camera to achieve accurate and reliable pose estimation. The framework is based on LiDAR-inertial system(LIS) and factor graph optimization theory. Texture information provided by vision is integrated into the LiDAR-inertial odometry to generate a new visual-inertial subsystem(VIS).The two subsystems, VIS and LIS, can assist each other and work jointly. Through real vehicle tests, the system can perform incremental, real-time state estimation, reconstruct dense 3D point cloud maps,and effectively solve the problems of localization drift and mapping failure in the lack of geometric features or challenging construction environments. Meanwhile, the system has a safety redundancy mechanism. When any subsystem fails, the system can also operate normally, to ensure the reliability and robustness of vehicle positioning.

The field of program classification, a critical aspect of software engineering, facilitates understanding and categorization of code across various applications, including anomaly detection, and code quality assessment. The advancement of cross-language program classification opens up avenues for efficient code translation among different programming languages, significantly aiding developers in rapid coding and reducing development cycles. Existing research primarily focuses on semantic code analysis, with limited emphasis on cross-linguistic challenges. To address this, we introduce an innovative neural network model, CodeSemanticsNN, which leverages a refined Syntax Structure (SS) approach. This model consists of two integral mechanisms: first, it adopts a novel SS representation that combines sequential and graph-based SS structures, enhancing semantic feature capture. Second, it employs a 'unified vocabulary' strategy to bridge the gap between diverse programming languages, facilitating efficient cross-language classification. Additionally, we have compiled a comprehensive dataset of 20,000 files spanning five programming languages, serving as a benchmark for cross-language classification. Our experiments on this dataset demonstrate that CodeSemanticsNN surpasses existing models in four key metrics: Precision, Recall, F1-score, and Accuracy.

Pepper leaf segmentation plays a crucial role in monitoring pepper leaf diseases in various backgrounds and ensuring the healthy growth of peppers. However, existing transformer-based segmentation methods suffer from computational inefficiency, excessive parameterization, and limited utilization of edge information. To tackle these challenges, we propose an adaptive multi-scale MLP framework, named AMS-MLP, which combines the multi-path aggregation module (MPAM) and the multi-scale context relation mask module (MCRD) to refine the object boundaries in pepper leaf segmentation. AMS-MLP consists of an encoder-based network, an adaptive multi-scale MLP (AM-MLP) module, and a decoder network. In the encoder network, the MPAM module effectively fuses five-scale features to generate a single-channel mask, improving the accuracy of pepper leaf boundary extraction. The AM-MLP module divides the input features into two branches: the global multi-scale MLP branch captures long-range dependencies between image information, while the local multi-scale MLP branch focuses on extracting local feature maps. Adaptive attention mechanism is designed to dynamically adjust the weights of global and local features. The decoder network incorporates the MCRD module into the convolutional layer, enhancing the extraction of boundary features. To verify the performance of the proposed method, we conducted extensive experiments on three pepper leaf datasets with different backgrounds. The results demonstrate mIoU scores of 97.39%, 96.91%, and 97.91%, as well as F1 scores of 98.29%, 97.86%, and 98.51%, respectively. Comparative analysis with U-Net and state-of-the-art models reveals that the proposed method dramatically improves the accuracy and efficiency of pepper leaf image segmentation.

The root system plays an irreplaceable role in plant growth. Its improvement can increase crop productivity. However, such system is still mysterious for us. The underlying mechanism has not been fully uncovered. The investigation on proteins related to the root system is an important means to complete this task. In the previous time, lack of root-related proteins makes it impossible to adopt machine learning methods for designing efficient models for the discovery of novel root-related proteins. Recently, a public database on root-related proteins was set up and machine learning methods can be applied in this field. In this study, we proposed a machine learning based model, named Graph-Root, for identification of root-related proteins. The features derived from protein sequences and one network were extracted, where the former features were processed by graph convolutional neural network and multi-head attention, and the later features abstracted the linkage between proteins. These features were fed into the fully connected layer to make prediction. The 5-fold cross-validation and independent tests suggested its good performance. It also outperformed the only one previous model, SVM-Root. Furthermore, the importance of each feature type and component in the proposed model was investigated.

Bearing fault diagnosis is crucial for ensuring the stable operation of mechanical equipment. With the continuous development of deep learning technology, Convolutional Neural Networks (CNNs) have demonstrated significant advantages in the field of fault diagnosis. This paper proposes a new method that combines various CNN architectures to improve the accuracy of bearing fault diagnosis. We designed five different convolutional network structures, including SerConv, ResConv, One-Shot Aggregation Convolution (OSAConv), Cross-Stage Aggregation Convolution (CSAConv), and MD-DAConv. Experimental results on the Case Western Reserve University (CWRU) bearing dataset show that the proposed method exhibits high accuracy and robustness in fault diagnosis. The results indicate that strategies such as multi-directional, multi-scale, and residual connections play a crucial role in enhancing the depth and breadth of feature extraction, while simple and effective feature fusion and information transmission mechanisms are key to ensuring the robustness and generalization ability of the model.

Use of a multi-sensor approach can provide citizens a holistic insight in the air quality in their immediate surroundings and assessment of personal exposure to urban stressors. Our work, as part of the ICARUS H2020 project, which included over 600 participants from 7 European cities, discusses data fusion and harmonization on a diverse set of multi-sensor data streams to provide a comprehensive and understandable report for participants, and offers possible solutions and improvements. Harmonizing the data streams identified issues with the used devices and protocols, such as non-uniform timestamps, data gaps, difficult data retrieval from commercial devices, and coarse activity data logging. Our process of data fusion and harmonization allowed us to automate the process of generating visualizations and reports and consequently provide each participant with a detailed individualized report. Results showed that a key solution was to streamline the code and speed up the process, which necessitated certain compromises in visualizing the data. A thought-out process of data fusion and harmonization on a diverse set of multi-sensor data streams considerably improved the quality and quantity of data that a research participant receives. Though automatization accelerated the production of the reports considerably, manual structured double checks are strongly recommended.

Unmanned Aerial Vehicles (UAVs) in particular multirotors are becoming the {\it de facto} tool for aerial sensing and remote inspection. In large industrial facilities, a UAV can transmit an online video stream to inspect difficult to access structures, such chimneys, deposits and towers. However, the communication range is limited, constraining the UAV operation range. This limitation can be overcome with relaying UAVs placed between the source UAV and the control station, creating a line of communication links. In this work we assume the use of a digital data packet network technology, namely WiFi, and tackle the problem of defining the exact placement for the relaying UAVs that creates an end-to-end channel with maximal delivery of data packets. We consider asymmetric communication links and we show an increase as large as $15$\% in end-to-end packet delivery ratio when compared to an equidistant placement. We also discuss the deployment of such a network and propose a fully distributed method that converges to the global optimal relay positions taking, on average, 1.4 the time taken by a centralized method.

Drug repositioning (also called drug repurposing) is a strategy for identifying new therapeutic targets for existing drugs. This approach is of great importance in pharmacology as it is a faster and cheaper way to develop new medical treatments. In this paper, we present, to our knowledge, the first application of multiplex-heterogeneous network embedding to drug repositioning. Network embedding learns the vector representations of nodes, opening the whole machine learning toolbox for a wide variety of applications including link prediction, node labelling or clustering. So far, the application of network embedding for drug repositioning focused on heterogeneous networks. Our approach for drug repositioning is based on multiplex-heterogeneous network embedding. Such method allows the richness and complexity of multiplex and heterogeneous networks to be projected in the same vector space. In other words, multiplex-heterogeneous networks aggregate different multi-omics data in the same network representation. We validate the approach on a task of link prediction and on a case study for SARS-CoV2 drug repositioning. Experimental results show that our approach is highly robust and effective for finding new drug-target associations.

Rollover prevention of partially-filled tank trucks is an industry teaser, with the core challenge being real-time perception and observation of the liquid state inside the tank. In order realize re-liable observation of sloshing liquid, this article firstly proposes a sloshing modeling method based on multi-degree-of-freedom pendulum model, and derives the double mass trammel pendulum model (DMTP, 2DOF) accordingly, which accurately reflects the sloshing dynamics under more pervasive working conditions. Secondly, a free surface fluctuation sensor is designed based on magnetostriction, capable of measuring the inclination and height of the liquid level inside tanks filled with hazardous chemical. Finally, the unscented Kalman filter (UKF) is utilized to synthesize the information of the two, establishing a credible real-time observation of the sloshing liquid. Verified through vehicle-fluid coupled co-simulation, under condition of con-secutive double lane change, observation error of the proposed method is only 25.9% that of the open-loop calculation, providing a secure guarantee for the observation of the state variables of the single pendulum model (SP) used for most kinds of anti-rollover controller.

We present a data-driven, in situ proximal multi-sensor digital soil mapping approach to develop digital twins for multiple agricultural fields. A novel Digital Soil CoreTM (DSC) Probe was engineered that contains seven sensors, each of a distinct modality, including sleeve friction, tip force, dielectric permittivity, electrical resistivity, soil imagery, acoustics, and visible and near-infrared spectroscopy. The DSC System integrates components the DSC Probe, DSC software, and deployment equipment to sense soil characteristics at a high vertical spatial resolution (mm scale) along in situ soil profiles up to a depth of 120 cm in about 60 sec. The DSC Probe in situ proximal data are harmonized into a data cube providing vertical high-density knowledge associated with physical-chemical-biological soil conditions. In contrast, conventional ex situ soil samples derived from soil cores, soil pits, or surface samples analyzed using laboratory and other methods are bound by substantially coarser spatial resolution and multiple compounding errors. Our objective was to investigate the effects of mismatched scale between high-resolution in situ proximal sensor data and coarser resolution ex situ soil laboratory measurements to develop soil prediction models. Our study was conducted in central California soil in almond orchards. We collected DSC sensor data and spatially co-located soil cores that were sliced into narrow layers for laboratory-based soil measurements. Partial Least Squares Regression (PLSR) cross-validation was used to compare results testing four data integration methods. Method A reduced the high-resolution sensor data to discrete values paired with layer-based soil laboratory measurements. Method B used stochastic distributions of sensor data paired with layer-based soil laboratory measurements. Method C allocated the same soil analytical data to each one of the high-resolution multi-sensor data within a soil layer. Method D linked the high-density multi-sensor soil data directly to crop responses (crop performance and behavior metrics) bypassing costly laboratory soil analysis. Overall, the soil models derived from Method C outperformed Methods A and B. Soil predictions derived using Method D were most cost-effective for directly assessing soil-crop relationships, making this method well-suited for industrial-scale precision agriculture applications.

：Non-Euclidean data, such as social networks, and citation relationships between documents, has node information and structural information. Graph Convolutional Network(GCN) can automatically learn node features and association information between nodes. The core ideology of the graph convolutional network is to aggregate node information by using edge information, thereby generating a new node feature. In the process of updating node features, there are two core influencing factors. One is the number of neighboring nodes of the central node, the other is the contribution of the neighboring nodes to the central node. Due to the previous GCN methods not simultaneously considering the numbers and different contributions of neighboring nodes to the central node, we design the Adaptive Attention Mechanism(AAM). To further enhance the representational capability of the model, we utilize Multi-Head Graph Convolution(MHGC). Finally, we adopt cross-entropy(CE) loss function to describe the difference between the predicted results of node categories and the ground truth (GT). Combined with backpropagation, this ultimately achieves accurate node classification. Based on AAM, MHGC and CE, we contrive the novel Graph Adaptive Attention Network (GAAN). Experiments show that the classification accuracy has achieved outstanding performances on Cora, Citeseer and Pubmed datasets.

Water networks are critical infrastructure components, ensuring the continuous supply of high-quality drinking water to consumers. To secure such water supply, regular maintenance, including the replacement of deteriorating pipelines, is essential. In this study, a methodology has been developed for determining optimal pipeline replacement solutions in water supply systems at water utilities with limited data availability. Hydraulic analysis has been conducted on the segment of 25 km of the water supply network using the free software EPANET. Applying water network optimization, eight pipeline replacement projects according to 13 pre-defined criteria have been identified and evaluated. The paper outlines the methods for evaluating the criteria, including defining specific quantitative limits. The Analytical Hierarchical Process (AHP) method was used in the paper to determine the weights of the criteria. The reason for applying this method refers to problems that involve a set of criteria with a mixed structure, including both quantitative and qualitative aspects. Also, the paper describes the steps of the multi-criteria optimization method VIKOR, used to select the optimal project. This paper, considered as case studie, describes a method, i.e. application of a new principles and an innovative way to solve a problem for developing countries.

Unmanned delivery technology using unmanned aerial vehicles (UAVs) has the potential to solve issues such as traffic congestion and labor shortages. We investigated a method to construct a route between the source and destination nodes in a wireless multi-hop network by installing a radio terminal in each house. If the relay functions of all nodes in the network are in operation, the number of control messages for route construction increases, and the message relay forwarding load of each node increases. Keeping all node pairs multi-hop connectivity, we establish a method to operate the relay function with as few nodes as possible. End nodes that have not activated the relay function connect to the relay node with the closest physical distance. In this paper, we propose methods for selecting nodes to operate the relay function. Specifically, one method is based on random selection, another method is based on the number of adjacent nodes, and a third method is based on the Multipoint Relay (MPR) of the Optimized Link State Routing Protocol (OLSR). For each method, we evaluate whether the route construction is available, the number of relay nodes, the total distance of constructed routes, and the distance in population-dense regions. We clarify the characteristics of each method, and the method based on MPR is the best for constructing a network suitable for UAV navigation routes. In addition, the smaller the number of relay nodes, the more unexpected collisions on the links due to route conflicts. We clarify the minimum node density required to apply the MPR method which reduces the number of relay nodes suppressing route conflicts.

Land cover mapping provides spatial information on the physical properties of the Earth’s surface, for various classes of wetlands, artificial surface and constructions, vineyards, water bodies, etc. Having reliable information on land cover is crucial to developing solutions to a variety of environmental problems such as destruction of important wetlands/forests, and loss of fish and wildlife habitats. This has made land cover mapping one of the most widespread application areas in remote sensing computational imaging. However, due to the differences between modalities in terms of resolutions, content, and sensors, integrating complementary information that multi-modal remote sensing imagery exhibits into a robust and accurate system still remains challenging, and classical segmentation approaches generally do not give satisfactory results for land cover mapping. In this paper, we propose a novel dynamic deep network architecture, AMM-FuseNet, that promotes the use of multi-modal remote sensing images for the purpose of land cover mapping. The proposed network exploits the hybrid approach of the Channel Attention mechanism and Densely Connected Atrous Spatial Pyramid Pooling (DenseASPP). In the experimental analysis, in order to to verify the validity of the proposed method, we test AMM-FuseNet applied to four datasets whilst comparing it to the 6 state-of-the-art models of DeepLabV3+, PSPNet, UNet, SegNet, DenseASPP, and DANet. In addition, we also demonstrate the capability of AMM-FuseNet under minimal training supervision (reduced number of training samples) compared to the state-of-the-art, achieving less accuracy loss even for the case with 1/20 of the training samples.

Often, engineers with machining experience often judge machining state and tool life according to chips’ features. Engineers' experience is digitized in this study. During the cutting process, the cutting tool coming in contact with the workpiece produces a shear zone, which causes plastic deformation and shear slip. The chips closest to the shear zone can directly show the state of the tool and workpiece when the material is SKD61. This study used chip color, vibration, and current signal integration for prediction of machining state and cutting tool life. When the cutting tool wears increased, the chip surface color changed in the following way: purpleè purple blueè blue ècyan, or even green and yellow. When the cutting tool was in the accelerating wear phase, the color change was particularly obvious. The Back-Propagation Levenberg–Marquardt (BP-LM) predictive methodology was used to compare the predictive ability of voltage, vibration signal, and chip color. The Mean Absolute Percentage Error (MAPE) for the voltage signal was 12.28%, for the vibration signal it was 11.38%, and for the chip color combined with multi-sensor characteristics it was 7.85%. The MAPE of the chip color was the smallest. Using the General Regression Neural Network (GRNN) methodology, the MAPE for the voltage signal was 10.74%, for the vibration signal 7.96%, and for the chip color combined with multi-sensor characteristics was 6.59%. The MAPE of the chip color was the smallest. Obviously, the chip color combined with multi-sensor signals provided better predictive results than the vibration signal or voltage signal alone. There is currently no research on the usefulness of monitoring chip color for tool life prediction.

With the rapid development of Intelligent Transportation Systems (ITS), vehicle type classification, as a key link in Automatic Toll Collection systems (ATC), is of great significance in improving traffic efficiency and reducing economic losses. This study proposes an intelligent vehicle classification system based on deep learning and multi-sensor data fusion to address the accuracy issues existing in vehicle classification methods based on optical sensors (OS) and human observers. The system significantly improves the accuracy and robustness of vehicle classification by combining deep Convolutional Neural Networks (CNN), LiDAR sensors, and machine learning algorithms. We first constructed a large-scale annotated dataset containing multiple vehicle types and complex traffic scenes to improve our model's capability to identify different vehicle characteristics. Next, CNN models based on different architectures were designed to extract global and local features of the vehicle, respectively. In addition, the LiDAR sensor was used to achieve the spatial structure architectures of the vehicle and combined with the output of the CNN model to improve classification performance under occlusion and complex scenes.

In this paper, an ultracompact combined sensor for displacement and angle synchronous measurement is proposed based on self-imaging effect of optical microgratings. Using a two-grating structure, linear and angular displacement can be measured by detecting the change of phase and amplitude of the optical transmission respectively within one single structure in meantime. The optical transmitted properties of the two-grating structure are investigated in both theory and simulation. Simulated results indicate that, the optical transmission changes in a sinusoidal relationship to the input linear displacement. Meanwhile, the amplitude of the curve decreases with an input pitch angle, indicating the ability for synchronous measurement within one single compact structure. The synchronous measurement of the linear displacement and the angle is also demonstrated experimentally. The results show a resolution down to 4nm for linear displacement measurement, and a sensitivity of 0.26mV/arcsec within a range of ±1° for angle measurement. Benefitting from a simple common-path structure without using optical components including reflectors and polarizers, the sensor shows an ultrahigh compactness for multiple-degrees-of-freedom measuring, indicating the great potential for this sensor in the fields such as integrated mechanical positioning and semiconductor fabrication.

With the development of deep learning, several Graph Neural Networks (GNN)-based approaches have been utilized for text classification. However, GNNs encounter challenges in capturing contextual text information within a document sequence. To address this, a novel text classification model RB-GAT is proposed by combining RoBERTa-BiGRU embedding and a multi-head Graph ATtention Network (GAT). First, the pre-trained RoBERTa model is exploited to learn word and text embeddings in different contexts. Second, the Bidirectional Gated Recurrent Unit (BiGRU) is employed to capture long-term dependencies and bidirectional sentence information from the text context. Next, the multi-head graph attention network is applied to analyze this information, which serves as a node feature for the document. Finally, the classification results are generated through a Softmax layer. Experimental results on three benchmark datasets demonstrate that our method can achieve an accuracy of 71.48%, 98.45%, and 80.32% on Ohsumed, R8, and MR, which is superior to the existing nine text classification approaches.

This study involves the development of a hybrid forecasting framework that integrates two different models in a framework to improve prediction capability. Although the concept of hybridization is not a new issue in forecasting, our approach presents a new structure that combines two standard simple forecasting models uniquely for superior performance. Hybridization is significant for complex data sets with multiple patterns. Such data sets do not respond well to simple models, and hybrid models based on the integration of various forecasting tools often lead to better forecasting performance. The proposed architecture includes serially connected ARIMA and ANN models. The original data set is first processed by ARIMA. The error (i.e., residuals) of the ARIMA is sent to the ANN for secondary processing. Between these two models, there is a classification mechanism where the raw output of the ARIMA is categorized into three groups before they are sent to the secondary model. The algorithm is tested on well-known forecasting cases from the literature. The proposed model performs better than existing methods in most cases.

In order to perform multi-degree-of-freedom motion measurements of marine machinery such as ship-borne mechanical platform in an absolute environment without a reference, absolute measurement methods using acceleration sensors and tilt gyroscopes are typically employed. However, due to the influence of wave forces on ship-borne mechanical platform, the coupling between different degrees of freedom can cause mutual interference, resulting in significant sensor measurement disturbances that make efficient computation and real-time analysis challenging. Specifically, when ship-borne mechanical platform swings, the proof mass of the acceleration sensor produces an undesirable signal output in the vertical direction, which leads to imprecise acceleration integration, thereby affecting accurate motion collection and posture estimation. To address these challenges, by analyzing the influence of the inclination angle of the ship-borne mechanical platform on the sensor measurement, and using the working principle of the acceleration sensor and angle sensor, a correction method for the motion measurement of the ship-borne mechanical platform based on multi-sensor fusion is proposed. In this article, we first analyze the influence of inclination angle on the integral effect in the heave direction. Then the configuration using four groups of acceleration sensors to correct the integral effect is proposed. Finally, the optimal inclination angle is determined through Kalman filtering based on the measured value of angle sensors and estimated values from the acceleration sensor sets. Experiments have proved that the average error of the corrected heave displacement signal is 25.34 mm, which is better than the integral displacement signal of a single acceleration sensor. At the same time, the acceleration sensor is used to calculate the roll angle and pitch angle of the ship-borne mechanical platform, and combined with the angle sensor signal to perform Kalman filtering, which filters out the errors caused by the shaking and instability of the mechanical platform, and can more accurately estimate the true inclination of the platform. Therefore, this method can enhance the precision and accuracy of the ship-borne mechanical platform motion signal acquisition, providing more valuable experimental data for research in marine engineering and related fields.

This paper focuses on robust multi-criteria traffic network equilibrium problems with path capacity constraints. The concepts of robust vector equilibrium and robust vector equilibrium with respect to worst case are introduced, respectively. To generate a subset of robust vector equilibrium, an equivalent min-max optimization problem is constructed. In addition, a direct search algorithm, in which the step size without derivatives and redundant parameters, for solving this min-max problem is proposed. To find a subset of robust vector equilibrium with respect to worst case, an equivalent smoothing optimization problem based on a variant version of ReLU activation function is established. Numerical experiments are also presented to expound the advantages of the algorithms.

This paper details the evolution of access network sharing models from legacy DSL to the most recent fibre-based technology and the main challenges faced from a technical and business perspectives. We first give an overview of existing access sharing models, that span physical local loop unbundling and virtual unbundled local access. We then describe different types of optical access technologies and highlight how they support network sharing. Next, we examine how the concept of SDN and network virtualization has been pivotal in enabling the idea of “true multi-tenancy”, through the use of programmability, flexible architecture and resource isolation. We give examples of recent developments of cloud central office and OLT virtualization. Finally, we provide an insight into the role that novel business models, such as blockchain and smart contract technology, could play in 5G networks. We discuss how these might evolve, to provide flexibility and dynamic operations that are needed in the data and control planes.

Aim: To determine whether an AI model and single sensor measuring acceleration and ECG could model cognitive and physical fatigue for a self-paced trail run. Methods: A field-based protocol of continuous fatigue repeated hourly induced physical (~45 minutes) and cognitive (~10 minutes) fatigue on one healthy participant. Physical load was a 3.8 km, 200 m vertical gain, trail run with acceleration and electrocardiogram (ECG) data collected using a single sensor. Cognitive load was a Multi Attribute Test Battery (MATB) and separate assessment battery including the Finger Tap Test (FTT), Stroop, Trail Making A and B, Spatial Memory, Paced Visual Serial Addition Test (PVSAT), and a vertical jump. A fatigue prediction model was implemented using a Convolutional Neural Network (CNN). Results: When the fatigue test battery results were compared for sensitivity to the protocol load, FTT right hand (R2 0.71) and Jump Height (R2 0.78) were the most sensitive while the other tests were less sensitive (R2 values Stroop 0.49, Trail Making A 0.29, Trail Making B 0.05, PVSAT 0.03, spatial memory 0.003). Best prediction results were achieved with a rolling average of 200 predictions (102.4 s), during set activity types, mean absolute error for ‘walk up’ (MAE200 12.5%) and range of absolute error for ‘run down‘ (RAE200 16.7%). Conclusion: We were able to measure cognitive and physical fatigue using a single wearable sensor during a practical field protocol including contextual factors in conjunction with a neural network model. This research has practical application to fatigue research in the field.

Device to Device (D2D) is a pivotal technology in the next generation of communication, allowing for direct task offloading between mobile devices (MDs) to improve the efficient utilization of idle resources. This paper proposes a novel algorithm for dynamic task offloading between active MDs and idle MDs in the D2D-MEC (Mobile Edge Computing) system by deploying multi-agent deep reinforcement learning (DRL) to minimize the long-term average delay of delay-sensitive tasks under deadline constraints. Our core innovation is a dynamic partitioning scheme for idle and active devices in the D2D-MEC system, accounting for stochastic task arrivals and multi-time-slot task execution, which has been insufficiently addressed in existing literature. We adopt a queue-based system to formulate a dynamic task offloading optimization problem. To address the challenges of large action space and coupling of actions across time slots, we model the problem as a Markov Decision Process (MDP) and perform the multi-agent DRL through Multi-Agent Proximal Policy Optimization (MAPPO). We employ a centralized training with decentralized execution (CTDE) framework to enable each MD to make offloading decisions solely based on its local system state. Extensive simulations demonstrate the efficiency and fast convergence of our algorithm. In comparison with the existing sub-optimal results deploying single-agent DRL, our algorithm reduces the average task completion delay by 11.0% and the ratio of dropped tasks by 17.0%.

Speech emotion recognition (SER) is important in facilitating natural human-computer interactions. In speech sequence modelling, a vital challenge is to learn context-aware sentence expression and temporal dynamics of para-linguistic features to achieve unambiguous emotional semantic understand-ing. In previous studies, the SER method based on the single-scale cascade feature extraction module could not effectively preserve the temporal structure of speech signals in the deep layer, downgrading the sequence modeling performance. In this paper, we propose a novel multi-scale feature pyramid network to mitigate the above limitations. With the aid of the bi-directional feature fusion of the pyramid network, the emotional representation with adequate temporal semantics is obtained. Experiments on the IEMOCAP corpus demonstrate the effectiveness of the proposed methods and achieve competitive results under speaker-independent validation.

Target detection and segmentation in synthetic aperture radar (SAR) images are vital steps for many remote sensing applications. In the era of data-driven deep learning, this task is extremely challenging due to the limited labeled data. Few-shot Learning has the ability to learn quickly from few samples with supervised information. Inspired by this, a few-shot learning framework named MSG-FN is proposed to solve the segmentation of ship targets in heterologous SAR images with few annotated samples. The proposed MSG-FN adopts a dual-branch network consisting of a support branch and a query branch. The support branch is used to extract features with an encoder, and the query branch uses a U-shaped encoder-decoder structure to segment the target in the query image. The encoder of each branch is composed of well-designed residual blocks combined with filter response normalization to capture robust and domain-independent features. A multi-scale similarity guidance module is proposed to improve the scale adaptability of detection by applying hand-on-hand guidance of support features to query features of various scales. In addition, a SAR dataset named SARShip-4i is built to evaluate the proposed MSG-FN and the experimental results show that the proposed method achieves superior segmentation results compared with the state-of-the-arts.

Accurate assessment of the breast cancer deterioration degree plays a crucial role in making medical plan, and the important basis for degree assessment is the number of mitoses in a given area of the pathological image. We utilized deep multi-scale fused fully convolutional neural network (MFF-CNN) combing with conditional random felid (CRF) to detect mitoses in hematoxylin and eosin stained histology image. Analyze the characteristics of mitotic detection ----scale invariance and sparsity, as well as the difficulties ---- small amount of data , inconsistent image staining and sample class unbalanced. Based on this, mitotic detection model is designed. In this paper, a tissue-based staining equalization method is used, and to establish an effective training sample set, we select training samples by using CNN. A mitotic detection model fusing multi-level and multi-scale features and context information was designed, and the corresponding training strategy was made to reduce over-fitting. As preliminarily validated on the public 2014 ICPR MITOSIS data, our method achieves a better performance in term of detection accuracy than ever recorded for this dataset.

Image segmentation is the process of partitioning an image into multiple segments where the goal is to simplify the representation of the image and make the image more meaningful and easier to analyze. In particular, semantic segmentation is an approach of detecting the classes of objects, based on each pixel. In the past, most semantic segmentation models were for only one single style, such as urban street views, medical images, or even Manga. In this paper, we proposed a semantic segmentation model called Dual-Pyramid Wide Residual Network (DPWRN) to solve the segmentation for cross-style datasets, which is suitable for diverse segmentation applications. The DPWRN integrated the Pyramid of Kernel paralleled with Dilation (PKD) and Multi-Feature Fusion (MFF) to improve the accuracy of segmentation. To evaluate the generalization of the DPWRN and its superiority over most state-of-the-art models, three datasets with completely different styles are tested in the experiments. Finally, the DPWRN verified its generalization and also showed its superiority over most state-of-the-art models.

When the image quality is evaluated, the human visual system (HVS) infers the details in the image through its internal generative mechanism. In this process, the HVS integrates both local and global information of the image, utilizes contextual information to restore the original image information, and compares it with the distorted image information for image quality evaluation. Inspired by this mechanism, a no-reference image quality assessment method is proposed based on a multitask image restoration network. The multitask image restoration network generates a pseudo-reference image as the main task and produces structural similarity index measure map as an auxiliary task. By mutually promoting the two tasks, a higher quality pseudo-reference image is generated. In addition, when predicting the image quality score, both the quality restoration features and the difference features between the distorted and reference images are used, thereby fully utilizing the information from the pseudo-reference image. To enable the model to focus on both global and local features, a multi-scale feature fusion module is proposed. Experimental results demonstrate that the proposed method achieves excellent performance on both synthetically and authentically distorted databases.

In the mold machining process, the cutting tool is worn with machining time, thereby affecting the surface accuracy, leading to poor workpiece dimensions, even fracture. At present, many studies have used multiple sensors to detect the machining conditions of cutting tool and workpiece, including indirect measurement method and direct measurement method. The indirect measurement method, which has been studied widely, mainly uses sensors to capture signals for subsequent data analysis; the direct measurement method mainly analyzes the state of cutting shear zone. Due to the cut-in of cutting tool in the machining process, the workpiece is dislocated rapidly, generating considerable amount of heat, which is transferred to the chips, inducing color change on the surface of chips. Many engineers with machining experience often judge the machining state and tool life according to the chips. The engineers' experience is digitized in this study, and indirect measuring sensors are used to predict the tool life, so as to attain the objective for smart manufacturing, the average percentage error of MAPE using single vibration and voltage eigenvalues as input features is 10%, the voltage signal characteristic values and vibration signal characteristic values are combined. Finally, the chip surface chromaticity eigenvalue is combined with signal characteristic value. The average prediction error of BP-LM method is 7.85%, the average prediction error of GRNN method is 6.59%. Therefore, when the eigenvalue of chip surface chromaticity is added to the prediction result, it can enhance the accuracy of cutting tool wear value prediction more effectively than single sensor signal characteristic value.

A stable driver state is essential in the process of manually transition control, which inevitably occurs in Level 3 automated driving situations. To this end, this paper proposed a CNN algorithm-based driver state monitoring system that uses multi-sensor data such as driver's face image, biometric information, and vehicle behavior information as input. This system calculates the probability of drowsiness for each of the four time periods using a convolutional neural network (CNN) based on ToF camera-based eye blinking, ECG information (pulse rate) embedded in the steering wheel, and vehicle information (steering angle data). In order to build a reliable and high-quality training dataset (Ground Truth) for the CNN algorithm, a baseline was established by matching the driver's face image with the electrocardiogram (ECG) and electroencephalogram (EEG) changes in the drowsy and normal states. In a simulation test of the proposed CNN algorithm using more than 20,000 driver image data acquired using a driving simulator, the TNR was 94.8% and the accuracy was 94.2%. Our proposed method is expected to minimize human errors that may occur when switching control by monitoring inappropriate driver state (drowsiness) in real time.

Image fusion is the process of combining multiple input images from single or multiple imaging modalities into a fused image, which is expected to be more informative for human or machine perception as compared to any of the input images. In this paper, we propose a novel method based on deep learning for fusing infrared images and visible images, named the LBP-based proportional input generative adversarial network (LPGAN). In the image fusion task, the preservation of structural similarity and image gradient information is contradictory, and it is difficult for both to achieve good performance at the same time. To solve this problem, we innovatively introduce Local Binary Patterns (LBP) into Generative Adversarial Networks (GANs), which effectively utilize the texture features of the source images, so that the network has stronger feature extraction ability and anti-interference ability. In the feature extraction stage, we introduce a pseudo-siamese network for the generator to extract the detailed features and the contrast features. At the same time, considering the characteristic distribution of different modal images, we propose a 1:4 scale input mode. Extensive experiments on the publicly available TNO dataset and CVC14 dataset show that the proposed method achieves the state-of-the-art performance. We also test the universality of LPGAN through the fusion of RGB and infrared images on the RoadScene dataset. In addition, LPGAN is applied to multi-spectral remote sensing image fusion. Both qualitative and quantitative experiments demonstrate that our LPGAN can not only achieve good structural similarity, but also retain rich detailed information.

Blade Health Monitoring (BHM) is often necessary in power plants and in aviation to prevent excessive blade vibration and cracks. This article proposes a network of blade tip timing sensors operated in a distributed BHM system. A number of cooperating agents is implemented in smart conditioning units which can autonomously operate in an adverse environment and communicate with other nodes via a serial interface. The project uses special versions of reduced instruction set chips that are able to operate near the hot section of the engine. Due to the limited number of types of microprocessors available in the extended temperature range grading, it was necessary to fully utilize the limited hardware resources and implement preemptive multitasking. For this purpose, a custom operating system and communication protocol were designed. The protocol hosts the middle layer which hides the implementation of the distributed system. The presented architecture ensures the sufficient computational capacity in individual nodes of the network operated in adverse conditions. It is scalable and resistant to transmission errors.

The robust fusion estimation problem of a multi-sensor system with communication constraints, random packet drops, d-step state delays, and deterministic control inputs is investigated. The state augmentation method is used to transform the time delay system into a non-time delay system, and then a robust fusion estimation algorithm based on sensitivity penalty is proposed and the pseudo-cross-covariance matrix is updated. An event-triggered robust fusion estimator with state delay and deterministic control inputs is considered to make the model closer to reality while effectively reducing network congestion and lowering communication costs. Conditions are given to ensure that the estimation errors of the robust fusion estimator are uniform boundedness. Two sets of numerical simulations are provided to illustrate the effectiveness of the derived fusion estimator.

Image segmentation is a new challenge prob- lem in medical application. The use of medical imaging has become an integral part of research, as it allows us to see inside the human body without surgical intervention. Many researcher have studied brain segmentation. One stage method is used to segment the brain tissues. In this paper, we proposed the multi-stage generative ad- versarial network to solve the problem of information loss in the one-stage. We utilize the coarse-to-fine to improve brain segmentation using multi-stage generative adversar- ial networks (GAN). In the first stage, our model generated a coarse outline for (i) background and (ii) brain tissues. Then, in the second stage, the model generated outline for (i) white matter (WM), (ii) gray matter (GM) and (iii) cerebrospinal fluid (CSF). A good result can be achieved by fusing the coarse outline and refine outline. We conclude that our model is more efficient and accu- rate in practice for both infant and adult brain segmenta- tion. Moreover, we observe that multi-stage model is faster than prior models. To be more specific, the main goal of multi-stage model is to see the performance of the model in a few shot learning case where a few labeled data are available. For medical image, this proposed model can work in a wide range of image segmentation where the convolution neural networks and one-stage methods have failed.

Traditional potential transformers have problems of large volume, difficulty in insulation, iron core saturation, ferroresonance overvoltage and poor transient response characteristics. The voltage sensor based on the principle of electric field coupling and differential input structure does not need to contact the measured object or ground, and can avoid the above problems. However, it requires a sufficiently high capacitance between the differential electrodes to obtain sufficient accuracy and a high voltage division ratio. The existing method of using mutual capacitance between the differential electrodes will cause many problems and fail to meet the practical needs. To solve the above problems, this paper innovatively uses multi-layer ceramic capacitor to replace the mutual capacitance and designs a new type of voltage sensor. In addition, by using single bypass small resistance grounding method to increase the input impedance of the differential signal processing circuit, error of the sensor is further reduced. The experimental results show that the sensor has excellent accuracy and great transient response characteristics. The ratio error under power frequency is within ±0.5% and the phase error is within 1. The ratio error in the range of 500 Hz∼30 kHz is within ±5% and the phase error is within 5. Moreover, it has the advantages of low cost, miniaturization, flexible shape and easy to adjust the voltage division ratio. These characteristics indicate that the sensor has good voltage measurement and sensor network potential.

In this paper, a multi-mode waveguide-based optical resonator is proposed for an integrated optical refractive index sensor. Conventional optical resonators have been studied for single-mode waveguide-based resonators to enhance the performance, but mass production is limited owing to the high fabrication costs of nano-scale structures. To overcome this problem, we designed an S-bend resonator based on a micro-scale multi-mode waveguide. In general, multi-mode waveguides cannot be utilized as optical resonators, because of a performance degradation resulting from modal dispersion and an output transmission with multi-peaks. Therefore, we exploited the mode discrimination phenomenon using the bending loss, and the resulting S-bend resonator yielded an output transmission without multi-peaks. This phenomenon is utilized to remove higher-order modes efficiently using the difference in the effective refractive index between the higher-order and fundamental modes. As a result, the resonator achieved a Q-factor and sensitivity of 2.3x10³ and 52 nm/RIU, respectively, using the variational finite-difference time-domain method. These results show that the multi-mode waveguide-based S-bend resonator with a wide line width can be utilized as a refractive index sensor.

Course recommendation is an important service of Intelligent tutoring systems (ITS), the rapid development of large-scale online courses has generated a substantial amount of learning process data for online learning platforms, providing valuable support for understanding learners' interests and preferences. However, despite made considerable achievements, there are still unresolved challenges:(1) Current research neglects the sequential relationship between the course learners learn. (2) It does not take advantage of the differences in learners' interest in different courses during the knowledge dissemination process. To overcome these challenges, we propose an innovative solution called PGDB (Preference-aware graph diffusion network and BiLSTM) for course recommendation. Specifically, we analyze learners' preferences for courses and relational paths using a relation-aware multi-head attention network, learn the semantic diversity of different contexts, effectively distinguish the difference of learners' interest in different courses in the process of knowledge transmission. In addition, the temporal preference modeling module employs a bi-directional long short-term memory network to mine learners' interest evolution patterns, generating learner-dependent representations, efficient use of the sequential relationship between the courses taken by learners. Experimental results on real datasets demonstrate that the proposed model greatly exceeds the performance of the most recent baseline models, thereby validating the model's effectiveness.

Blind deconvolution (BD) is one of the effective methods that help pre-process vibration signals and assist in bearing fault diagnosis. Currently, most BD methods design an optimization criterion and use frequency or time domain information independently to optimize a deconvolution filter. It recovers weak periodic impulses related to incipient faults. However, the random noise interference may cause the optimizer to overfit. The time-domain-based BD methods tend to extract fault-unrelated single peak impulse, and the frequency-domain-based BD methods tend to retain the maximum energy frequency component, which will lose the fault-related harmonics frequency components. To solve the above issue, we propose a hybrid criterion that combines the kurtosis for time domain optimization and the $G-l_1/l_2$ norm for the frequency domain. These two criteria are monotonically increasing and decreasing, so they mutually constrain to avoid overfitting. After that, we design a multi-task one-dimensional convolutional neural network with time and frequency branches to achieve an optimal solution for this hybrid criterion. The multi-task neural network realizes the simultaneous optimization of two domains. Experimental results show that our proposed method outperforms other state-of-the-art methods.

This work analyzes a new indoor laboratory data set looking at early fire indicators in controlled and realistic experiments representing different incipient fire scenarios. The experiments were performed within the confines of an indoor laboratory setting using multiple distributed sensor nodes on different room positions. Each sensor node collected data of particulate matter (PM), volatile organic compounds (VOC), CO, CO2, H2, UV, air temperature and humidity in terms of a multivariate time series. These data hold immense value for researchers within the machine learning and data science communities who are keen on exploring innovative advanced statistical and machine learning techniques. They serve as a valuable resource for the development of early fire detection systems. The analysis of the collected data was carried out in dependence of the manhattan distance between the fire source and the sensor node. We found that especially larger particles (> 0.5 μm) and VOCs show a significant dependency with respect to the intensity as a function of the manhattan distance to the source. Moreover, we observed differences in the propagation behavior of VOCs, PM, and CO, which are particularly relevant in incipient fire scenarios due to the presence of strand popagation effects.

With the latest advances in information and communication technologies, greater amounts of sensitive user and corporate information are constantly shared across the network making it susceptible to an attack that can compromise data confidentiality, integrity and availability. Intrusion Detection Systems (IDS) are important security mechanisms that can perform a timely detection of malicious events through the inspection of network traffic or host-based logs. Throughout the years, many machine learning techniques have proven to be successful at conducting anomaly detection but only a few considered the sequential nature of data. This work proposes a sequential approach and evaluates the performance of a Random Forest (RF), a Multi-Layer Perceptron (MLP) and a Long-Short Term Memory (LSTM) on the CIDDS-001 dataset. The resulting performance measures of this particular approach are compared with the ones obtained from a more traditional one, that only considers individual flow information, in order to determine which methodology best suits the concerned scenario. The experimental outcomes lead to believe that anomaly detection can be better addressed from a sequential perspective and that the LSTM is a very reliable model for acquiring sequential patterns in network traffic data, achieving an accuracy of 99.94% and a f1-score of 91.66%.

A Wireless Body Area Sensor Network (WBASN) is combination of numerous sense nodes, positioned onto/close or inside a person body. Wireless Body Area Sensor Networks (WBASN) is a developing automation trend that exploits wireless sensor nodes to put instantaneous wearable well-being of ill person to improve individual’s existence. The sensor nodes might be used outwardly to observe abundant health parameters (like heart activity, blood pressure and cholesterol) of an ill person at a vital site within hospital. Hence the goal of WBASN is much crucial, enhancing the lifetime of nodes is compulsory to sustain many issues such as utility and efficiency. It is essential to evaluate time that when the first node will die it we want to refresh or change the battery reason is that loss of crucial information is not tolerable. The lifetime is termed as the time interval when a first node dies out due to battery exhaustion. In our proposed protocol life time of a network is the main concern as well other protocol related issues such as throughput, path loss, and residual energy. Bio-sensors are used for deployment on human body. Poisson distribution and equilibrium model techniques have been used for attaining the required results. Multi-hop network topology and random network node deployment used to achieve minimum energy consumption and longer network lifetime.

Objective: Scaphoid fractures, particularly occult and non-displaced fractures, are difficult to detect using traditional X-ray methods due to their subtle appearance and variability in bone density. The study proposes a two-stage CNN approach to detect and classify scaphoid fractures using anterior-posterior (AP) and lateral (LA) X-ray views for more accurate diagnosis. Methods: The study emphasizes the use of multi-view X-ray images (AP and LA views) to improve fracture detection and classification. The multi-view fusion module helps integrate information from both views to enhance detection accuracy, particularly for occult fractures that may not be visible in a single view. The proposed method includes two stages that are stage 1: Detect the scaphoid bone using Faster R-CNN and Feature Pyramid Network (FPN) for region proposal and small object detection. The detection accuracy for scaphoid localization is 100% with Intersection over Union (IoU) scores of 0.8662 for AP views and 0.8478 for LA views. And stage 2: Perform fracture classification using a ResNet backbone and FPN combined with a multi-view fusion module to combine features from both AP and LA views. This stage achieves a classification accuracy of 89.94%, recall of 87.33%, and precision of 90.36%. Results: The proposed model performs well in both scaphoid bone detection and fracture classification. The multi-view fusion approach significantly improves recall and accuracy in detecting fractures compared to single-view approaches. In scaphoid detection, both AP and LA views achieved 100% detection accuracy. In fracture detection, using multi-view fusion, the accuracy for AP views reached 87.16%, and for LA views, it reached 83.83%. Conclusion: The multi-view fusion model effectively improves the detection of scaphoid fractures, particularly in cases of occult and non-displaced fractures. The model provides a reliable, automated approach to assist clinicians in detecting and diagnosing scaphoid fractures more efficiently.

Based on the topology of wide area protection system (WAPS), after studying the reliability of hardware system and information flow in the WAPS and establishing the reliability assessment model, the multi-service reliability analysis method with multi monitoring and protection tasks in WAPS was proposed. In the model, the impact of network quality of service (QoS) such as information flow loss and delay, is studied. On the base of the model, the multi-service reliability evaluation method is employed to analyze the reliability of a WAPS of IEEE14 node power system, and the key nodes of the WAPS is given, which provides a basis for improving the reliability of the WAPS.

Using traditional methods based on detection rules written by human security experts, there is a significant challenge in accurately detecting many network threats that are increasingly becoming sophisticated. In order to deal with the limitation of the traditional methods, network threat detection techniques utilizing artificial intelligence technologies such as machine learning are being extensively studied. Research has also been conducted on analyzing various string patterns in network packet payloads through natural language processing techniques to detect attack intent. However, due to the nature of packet payloads containing binary data as well as text data, a new approach is needed that goes beyond typical natural language processing techniques. In this paper, we study a token extraction method optimized for payloads using n-gram and byte pair encoding techniques. Furthermore, we generate embedding vectors that can understand the context of the packet payload using algorithms such as Word2Vec and FastText. We also compute the embedding of various header data associated with packets, such as IP addresses and ports. Given the combination of these features, we ensemble a text 1D-CNN and a multi-head attention network in a novel fashion. We validated the effectiveness of our classification technique with the CICIDS2017 open dataset and over half a million data collected by The Education Cyber Security Center (ECSC) currently operating in South Korea. The proposed model showed remarkable progress compared to previous studies, demonstrating a highly accurate classification performance with an F1-Score of 0.998. Our model can also preprocess and classify 150,000 network threats per minute to help security agents in the field divert their time to analyzing more complex attack patterns.

Internet of things (IoT) is a network of smart things. This indicates the ability of these physical things to transfer information with other physical things. IoT has introduced various services and daily human life depends on its reliable and accessible operation. The characteristics of these networks, such as topology dynamicity and energy constraint, challenges the routing problem in these networks. Previous routing methods could not achieve the required performance in this type of network. Therefore, developers of this network designed and developed specific methods in order to satisfy the requirements of these networks. One of the routing methods is utilization of multi-path protocols which send data to its destination using routs with separate links. One of such protocols is AOMDV routing protocol. AOMDV protocol is a multi-path protocol which uses multiple different paths for sending information in order to maintain the network traffic balance, manage and control node energy, decrease latency, etc. In this paper, this method is improved using gray system theory which chooses the best paths used for separate routes to send packets. To do this, AOMDV packet format is altered and some fields are added to it so that energy criteria, link expiration time, and signal to noise ratio can also be considered while selecting the best route. The proposed method named GSTMPR-IoT is introduced which chooses the routs with highest rank for concurrent transmission of data, using a specific routine based on the gray system theory. In order to evaluate and report the results, the proposed GSTMPR-IoT method is compared to the EECRP and AOMDV approaches with regard to throughput, packet delivery rate, end to end delay, average residual energy, and network lifetime. The results demonstrate the superior performance of the proposed GSTMPR-IoT compared to the EECRP and AOMDV approaches.

The progressive impairment analysis in gait from neurological diseases patients such as Hereditary Ataxias (HA) has been carried out using gait data collected with movement sensors. This research is focused on finding the minimum amount required of gait features to recognize efficiently and less intrusive way, HA patients based on data collected with iPhone movement sensors placed on the ankles from 14 HA patients and 14 healthy people. A twofold proposal is made , first a local minimum prominent peak criterion to find out the starting point of each stride, to get 10-stride window about which 56 spatial-temporal features are derived; second a search strategy based on Hill Climbing algorithm to reduce the number of gait features and sensors. The main results were the findings that with two gait patterns a 96% of classification accuracy was achieved by using K-Nearest Neighbors (KNN) and Multi-Layer Perceptron (MLP) algorithms, but in addition, MLP only right ankle sensor patterns were required which also allows to reduce the intrusion.

Underwater robotics equipped with visual detection systems have the function of detecting un-derwater artifacts, which is of great significance to deep-sea archaeology. Underwater artifacts are located in complex environments with poor imaging conditions, and the targets have chal-lenging problems such as breakage, stacking, and sediment burial, which causes the marine cul-tural target detection failures. To solve these problems, this paper proposes an underwater cul-tural target detection algorithm based on the deformable deep aggregation network model for underwater robotics exploration. To fully extract the target feature information of underwater targets in complex environments, this paper designs a multi-scale deep aggregation network with deformable convolutional layers. Besides, the BAM attention module is designed for fea-ture optimization, which enhances the potential feature information of the target while weak-ening the background interference information. Finally, the target prediction is achieved through feature fusion at different scales. The proposed algorithm has been extensively validat-ed and analyzed on the collected underwater artifact datasets, and the precision, recall, and mAP of the algorithm have reached 93.1%, 91.4%, and 92.8%, respectively. In addition, the present al-gorithm has been practically deployed on the underwater robotics. In the deep-sea tests, the ar-tifact detection frame rate reaches up to 19 fps, which satisfies the real-time target detection。

Accurate and timely prediction of Napa cabbage (Brissica rapa subsp. Perkinensis) fresh weight is crucial for optimizing harvest timing, crop management, and supply chain logistics, contributing to food security and price stabilization. Traditional manual sampling methods are labor-intensive and imprecise. This study addresses this challenge by developing a comprehensive (artificial intelligence) AI-powered model for predicting Napa cabbage fresh weight using unmanned aerial vehicle (UAV)-based multi-sensor data. High-resolution RGB, multispectral, and thermal infrared (TIR) imagery were collected over a Napa cabbage field throughout the 2020 growing season. Various vegetation indices, crop features (vegetation fraction, crop height model), and water stress indi-cators (CWSI) were extracted from the imagery. Three AI algorithms—deep neural network (DNN), support vector machine (SVM), and random forest (RF)—were trained and evaluated, with the DNN model consistently outperforming the others. The DNN model achieved the highest accuracy (R² = 0.86 for training, 0.82 for testing; root mean square error (RMSE) = 0.432 kg for training, 0.465 kg for testing) during the mid-to-late rosette growth stage (DAP 35-42), highlighting this period as crucial for fresh weight estimation due to stable leaf area and well-developed canopy structure. The model tended to underestimate the weight of Napa cabbages exceeding 5 kg, potentially due to limited samples and saturation effects of vegetation indices. However, the overall error rate was less than 5%, demonstrating the feasibility and effectiveness of this approach. Spatial analysis revealed that the model accurately captured the variability in Napa cabbage growth across different soil types and irrigation conditions, particularly reflecting the positive impact of drip irrigation on the sandy loam plot. Bias analysis indicated the DNN model's tendency to overestimate smaller Napa cabbages (2 kg), suggesting areas for future refinement. This study demonstrates the potential of UAV-based multi-sensor data and AI algorithms for accurate and non-invasive prediction of Napa cabbage fresh weight. The developed DNN model offers a promising tool for optimizing harvest timing, improving crop management practices, and en-hancing supply chain efficiency. Future research should focus on refining the model for specific weight ranges and diverse environmental conditions, as well as extending its application to other crops, to further advance precision agriculture and contribute to sustainable food production.

The design and usage of the addressed combined fiber-optic sensors (ACFOS) and the multisensory control systems of the greenhouse gas concentration on their basis are investigated. The main development trend of the combined fiber-optic sensors (CFOS), consisting of the fiber Bragg grating (FBG) and the Fabry-Perot resonator (FPR), which are successively formed at the optical fiber end, is highlighted. The addressed fiber Bragg structures (AFBS) usage instead of the FBG in the CFOS leads not only to significant cheapening of the sensor system due to microwave photonics interrogating methods, but also to increasing its metrological characteristics. The structural scheme of the multisensory gas concentration monitoring system is suggested. The suggested scheme allows detecting four types of the greenhouse gases (СО2, NO2, CH4, OX) depending on the material and thickness of the polymer film, which is the FPR sensitive element. The usage of Karunen-Loeff transform (KLT), which allows separating each component contribution to the reflected spectrum according to its efficiency, is proposed. In the future, it allows determining the gas concentration at the AFBS address frequencies. The estimations have shown that the ACFOS design in the multisensory system allows measuring the environment temperature in the range of −60…+300 °C with an accuracy of 0.1–0.01 °C, and the gas concentration in the range of 10…90% with the accuracy of 0.1–0.5%.

Pedestrian object detection plays a significant role in intelligent systems such as intelligent traffic and monitoring. Traditional machine learning methods on pedestrian detection have shown various drawbacks, e.g., low accuracy, slow speed, etc. The Convolutional Neural Network (CNN) based object detection algorithms have demonstrated remarkable advantages in the field of pedestrian detection. However, the mainstream CNNs still face the problems of slow speed and low detection accuracy, especially on small and occluded targets from aerial perspective. In this paper, we propose Multi-Scale Attention YOLO (MSA-YOLO) detection algorithm to address the above issues. MSA-YOLO includes a Squeeze, Excitation and Cross Stage Partial (SECSP) channel attention module for CNNs to extract richer pedestrian features with a small number of extra parameters. It also contains a multi-scale prediction module to capture the information among different pedestrian scales, which can recognize the small objects with higher accuracy and significantly reduce the missed detection. To sufficiently evaluate our proposed model, we manually collect and annotate a new benchmark dataset, Aerial Pedestrian Dataset, which has much more sample annotations, features, scenes and image view angles than the existing benchmark datasets. In addition, the images in our dataset have higher resolution than most of benchmark pedestrian detection datasets, which can provide more detailed features of pedestrians and thus improve the model performance. When tested on Aerial Pedestrian Dataset, our proposed MSA-YOLO algorithm significantly outperforms the most used baseline models with almost the same model size. This shows the efficiency of our proposed model. (The code and new dataset will be released to the public later.)

Internet of things (IoT) is a network of smart things. This indicates the ability of these physical things to transfer information with other physical things. The characteristics of these networks, such as topology dynamicity and energy constraint, challenges the routing problem in these networks. Previous routing methods could not achieve the required performance in this type of network. Therefore, developers of this network designed and developed specific methods in order to satisfy the requirements of these networks. One of the routing methods is utilization of multi-path protocols which send data to its destination using routes with separate links. One of such protocols is AOMDV routing protocol. In this paper, this method is improved using gray system theory which chooses the best paths used for separate routes to send packets. To do this, AOMDV packet format is altered and some fields are added to it so that energy criteria, link expiration time, and signal to noise ratio can also be considered while selecting the best route. The proposed method named RMPGST-IoT is introduced which chooses the routes with highest rank for concurrent transmission of data, using a specific routine based on the gray system theory. In order to evaluate and report the results, the proposed RMPGST-IoT method is compared to the ERGID and ADRM-IoT approaches with regard to throughput, packet receiving rate, packet loss rate, average remaining energy, and network lifetime. The results demonstrate the superior performance of the proposed RMPGST-IoT compared to the ERGID and ADRM-IoT approaches.

The Onion Route Network (also called TOR) is by far the most efficient and widely used anonymity platform with millions of users daily and an expanding size and capacities. Since its public deployment in 2002, the Onion Routing network (also known as TOR) has maintained its leading position and dozens of propositions aiming to improve its performance and enhance the security (anonymity and privacy) have been made. Given the significance of this research area, this work seek to contribute into the improvement of TOR by investigating and testing revolutionary cryptographic and routing mechanism. This work is justified by the current TOR vulnerability and observed weaknesses, and set the challenging aim of covering these security flaws by proposing the relevant security and performances improvement such as the authenticated-encryption for onion construction, the lightened onion encapsulation approach and the secured circuit selection and cell routing mechanisms. The obtained results from implementing the proposed improvements and testing them into a TOR-like simulation platform permit to validate not only, the performances and security contributions brought by the improvement, but also the suitability of their potential implementation into the real TOR network.

Deep learning classification is the state-of-the-art of machine learning approach. Earlier work proves that the deep convolutional neural network has successfully and brilliantly in different applications such as images or video data. Recognizing and clarifying the remote sensing aspect of the earth's surface and exploit land cover and land use (LCLU). First, this article summarized the remote sensing emerging application and challenges for deep learning methods. Second, we propose four approaches to learn efficient and effective CNNs to transfer image representation on the ImageNet dataset to recognize LCLU datasets. We use VGG16, Inception-ResNet-V2, Inception-V3, and DenseNet201 models to extract features from the EACC dataset. We use pre-trained CNNs on ImageNet to extract features. For feature selection we proposed principal component analysis (PCA) to improve accuracy and speed up the model. We train our model by multi-layer perceptron (MLP) as a classifier. Lastly, we apply the multi-granularity encoding ensemble model. We achieve an overall accuracy of 92.3% for the nine-class classification problem. This work will help remote sensing scientists understand deep learning tools and apply them in large-scale remote sensing challenges

This study presents a research plan that utilizes data obtained from wearable devices to identify human activities and gain insights into human behavior. We developed a model capable of classifying activities similar to human behavior and evaluated the effectiveness and generalization capabilities of this model. The data underwent initial preprocessing, including standardization and normalization. Additionally, recognizing the inherent similarities between human activity behaviors, we introduced a multi-layer classifier model. The first layer is a random forest model based on stepwise regression, which may encounter reduced accuracy for similar activities. The second layer employs a Support Vector Machine (SVM) model based on Kernel Fisher Discriminant Analysis (KFDA). KFDA is used to reduce the dimensionality of data points with potential confusion, followed by SVM for classification. The model was experimentally evaluated and applied to four benchmark datasets: UCI DSA, UCI HAR, WISDM, and IM-WSHA. The experimental results demonstrate that our approach achieved recognition accuracies of 99.71%, 98.71%, 99.12%, and 97.6% on these datasets, indicating excellent recognition performance. Furthermore, to assess the model's generalization ability, we performed K-fold cross-validation on the random forest model and utilized ROC curves for the SVM classifier. The results indicate that our multi-layer classifier model exhibits robust generalization capabilities.

In this paper, stochastic multi-objective intelligent framework (MOIF) is performed for distribution network reconfiguration to minimize power losses, the number of voltage sags, the system's average RMS fluctuation, the average system interruption frequency (ASIFI), the momentary average interruption frequency (MAIFI), and the system average interruption frequency (SAIFI) considering the network uncertainty. The unscented transformation (UT) approach is applied to model the demand uncertainty due to simplicity to implement and no presumptions to simplify. A human-inspired intelligent method named improved mountaineering team-based optimization (IMTBO) is applied to determine the decision variables defined as the network's optimal configuration. The conventional MTBO is improved using a quasi-opposition-based learning strategy to overcome premature convergence and achieve the optimal solution. The simulation results showed that in single- and double-objective optimization, some objectives are weakened compared to their base value, while the results of the MOIF indicated a fair compromise between different objectives, and all objectives are improved. The results of MOIF based on the IMTBO cleared that the losses are reduced by 30.94%, the voltage sag numbers and average RMS fluctuation are reduced by 33.68% and 33.65%, and also ASIFI, MAIFI, and SAIFI are improved by 6.80%, 44.61%, and 0.73%, respectively. Also, the superior capability of the MOIF based on the IMTBO is proved compared to the conventional MTBO, particle swarm optimization, and artificial electric field algorithm. Moreover, the results of the stochastic MOIF based on the UT showed the power loss increased by 7.62%, voltage sag, and SARFI increased by 5.39% and 5.31%, and ASIFI, MAIFI, and SAIFI weakened by 2.28%, 6.61%, and 1.48%, respectively compared to the deterministic MOIF model.

The presence of fog in the background can prevent small and distant objects from being detected, let alone tracked. Under safety-critical conditions, multi-object tracking models require faster-tracking speed while maintaining high object-tracking accuracy. The original DeepSORT algorithm used YOLOv4 for the detection phase, and a simple neural network for deep appearance descriptor. Consequently, the feature map generated loses relevant details about the track being matched with a given detection in fog. Targets with a high degree of appearance similarity on the detection frame are more likely to be mismatched, resulting in identity switches or track failures in heavy fog. We propose an improved multi-object tracking model based on the DeepSORT algorithm to im-prove tracking accuracy and speed under foggy weather conditions. First, we employed our camera-radar fusion network (CR-YOLOnet) in the detection phase for faster and more accurate object detection. We proposed an appearance feature network to replace the basic convolutional neural network. We incorporated GhostNet to take the place of the traditional convolutional layers to generate more features and reduce computational complexities and cost. We adopted a segmentation module and fed the semantic labels of the corresponding input frame to add rich semantic information to the low-level appearance feature maps. Our proposed method outperformed YOLOv5 + DeepSORT with a 35.15% increase in multi-object tracking accuracy, a 32.65% increase in multi-object tracking precision, the speed increased by 37.56%, and identity switches decreased by 46.81%.

Deep learning methods have been widely used in PolSAR image classification. To learn the polarimetric information, many deep learning methods expect to learn high-level semantic features from original PolSAR data. However, only original data cannot learn multiple scattering features and complex structures for extremely heterogeneous terrain objects. In addition, deep learning methods always cause edge confusion due to the high-level features. To overcome these shortages, we propose a double-channel CNN network combined with an edge-preserving MRF model(DCCNN-MRF) for PolSAR image classification. Firstly, to combine complex matrix data and multiple scattering features together, a double-channel convolution network(DCCNN) is developed, which consists of a Wishart-based complex matrix and multi-feature subnetworks. The Wishart-based complex matrix network can learn the statistical characteristics and channel correlation well, and the multi-feature network can learn high-level semantic features well. Then, a unified network framework is designed to fuse two kinds of features to enhance advantageous features and reduce redundant ones. Finally, an edge preserving MRF model is designed to combine with the DCCNN network. In the MRF model, a sketch map-based edge energy function is designed by defining adaptive weighted neighborhood for edge pixels. Experiments are conducted on four real PolSAR data sets with different sensors and bands. Experimental results demonstrate the effectiveness of the proposed DCCNN-MRF method.

This study analyses the changes in land use in the Sado estuary (West-Central Portugal) based on a multitemporal analysis of cartographic data from the 19th century and remote sensing land use maps from the 21st century, updated by field work. A GIS plotting of land-use evolution is summarized in a quantitative table. The comparison shows the changes in land use, with increasing occupation by human economic activities, including extensive agriculture and forestry, as well as localized urbanization and industrialization. The main elements of the landscape impacted by anthropogenic uses were: i) hydrography - river dams affected the flow dynamics and sedimentary processes in the estuary; ii) vegetation – increasing agriculture and forestry reduced the areas of native vegetation, nowadays mostly occupied by vineyards, pine forests and cork oaks. iii) wetlands - tidal and alluvial plains are being occupied by rice cultivation, aquaculture, industries, and ports; iv) coastal dunes – new developments are occupying large areas of Holocene coastal dunes; v) natural environment – mining and dredging affected some habitats and biodiversity. This analysis is intended to help the territorial organization of present and future economic activities, as well as to reduce environmental and social problems, promoting long-term sustainability of this rapidly evolving region.

The incorporation of electric vehicles (EVs) with vehicle-to-grid (V2G) offers the development of multi-energy microgrid (MMOs) models. However, MMOs have among their goals to decrease electricity costs by coordinating with vehicle-to-grid (V2G) scheduling. Controlling V2G models poses uncertainties about the voltages developed and creates the need for power generation disruptions. In this work, a Feed-Forward-Back-Propagation Network (FFBPN) is developed with Matlab software, based on the Levenberg-Marquardt algorithm by varying the number of hidden neurons to realize better performance thanks to the value of the Maximum Coefficient of De-termination (R2) and the Minimum Mean Squared Error (MSE). In this article, the dataset from NASA PCoE has been used to make proper consideration of the model that best calculates the life cycle of the battery, demonstrating relevant implications for future development of this type of system based on FFBPN for the integration of EV into V2G. The proposed FFBPN demonstrates better performance than other methods taken from the literature. In conclusion, the comparison between the phases of training, validation and test are used to completely represent the validity of the proposed model and precisely to identify the characteristics curves of FFBPN for future ap-plications highlighting its performance to create profitability, efficiency, production, energy saving and minimum environmental impact.

Remote sensing image change detection is a key application technology in remote sensing. In recent years, the development of deep learning has brought new opportunities for remote sensing image change detection. However, to achieve high change detection accuracy, large-scale, good-quality labeled data is required to train the model, which is often difficult to obtain and costly. This paper proposes a semi-supervised remote sensing image change detection network based on consistent regularization for multi-scale feature fusion and reuse (SCMFM-CDNet), which can enhance change detection performance with limited labeled data. Specifically, the SCMFM-CDNet consists of two parts, the supervised stage multi-scale feature fusion and reuse network (MFM-CDNet) and the consistency regularization method of semi-supervised learning. The MFM-CDNet network uses channel attention mechanism and spatial attention mechanism to enhance the expression ability of features, and gradually completes the extraction and detection of change features through multi-scale feature fusion and reuse strategy. The consistency regularization method of semi-supervised learning adopts the teacher model and student model for alternate training, updating model parameters by minimizing the differences in their predictions, thus achieving semi-supervised learning. This paper conducted extensive experiments on three public datasets, LEVIR-CD, WHU-CD, and GoogleGZ-CD. Under the three labeling data ratios of 5%, 10%, and 20%, the F1 scores of the SCMFM-CDNet network on the LEVIR-CD dataset were 88.62, 90.78, and 90.63, respectively. On the WHU-CD dataset, the F1 scores were 88.40, 90.17, and 90.83, respectively. On the GoogleGZ-CD dataset, the F1 scores were 81.44, 82.36, and 85.40, respectively. The results indicate that the proposed network outperforms existing semi-supervised change detection methods in terms of comprehensive performance. In addition, this paper also conducted ablation experiments to evaluate the contribution of the global context module (GCM) and feature fusion reuse module (FFM) in the MFM-CDNet network. The experimental results show that these two modules are crucial for improving network performance.

Despite numerous studies in statistical downscaling methodology, there remains a lack of methods that can downscale from precipitation modeled in global climate models to regional level high resolution gridded precipitation. This paper reports a novel downscaling method using a Generative Adversarial Network (GAN), CliGAN, which can downscale large-scale annual maximum precipitation given by simulation of multiple atmosphere-ocean global climate models (AOGCM) from Coupled Model Inter-comparison Project 6 (CMIP6) to regional-level gridded annual maximum precipitation data. This framework utilizes a convolution encoder-dense decoder network to create a generative network and a similar network to create a critic network. The model is trained using an adversarial training approach. The critic uses the Wasserstein distance loss function and the generator is trained using a combination of adversarial loss Wasserstein distance, structural loss with the multi-scale structural similarity index (MSSIM), and content loss with the Nash-Sutcliff Model Efficiency (NS). The MSSIM index allowed us to gain insight into the model’s regional characteristics and shows that relying exclusively on point-based error functions, widely used in statistical downscaling, may not be enough to reliably simulate regional precipitation characteristics. Further use of structural loss functions within CNN-based downscaling methods may lead to higher quality downscaled climate model products.

In the remote sensing domain, it is crucial to complete semantic segmentation on the raster images, e.g., river, building, forest, etc, on raster images. A deep convolutional encoder--decoder (DCED) network is the state-of-the-art semantic segmentation method for remotely sensed images. However, the accuracy is still limited, since the network is not designed for remotely sensed images and the training data in this domain is deficient. In this paper, we aim to propose a novel CNN for semantic segmentation particularly for remote sensing corpora with three main contributions. First, we propose applying a recent CNN called a global convolutional network (GCN), since it can capture different resolutions by extracting multi-scale features from different stages of the network. Additionally, we further enhance the network by improving its backbone using larger numbers of layers, which is suitable for medium resolution remotely sensed images. Second, "channel attention'' is presented in our network in order to select the most discriminative filters (features). Third, "domain-specific transfer learning'' is introduced to alleviate the scarcity issue by utilizing other remotely sensed corpora with different resolutions as pre-trained data. The experiment was then conducted on two given datasets: (i) medium resolution data collected from Landsat-8 satellite and (ii) very high resolution data called the ISPRS Vaihingen Challenge Dataset. The results show that our networks outperformed DCED in terms of $F1$ for 17.48% and 2.49% on medium and very high resolution corpora, respectively.

The production of lithium-ion battery cells is characterized by a high degree of complexit due to numerous cause-effect relationships between process characteristics. Knowledge about the multi-stage production is spread among several experts, rendering tasks such as failure analysis challenging. In this paper, a method is presented, which includes expert knowledge acquisition in production ramp-up by combining Failure Mode and Effects Analysis (FMEA) with a Bayesian Network. We show the effectiveness of this holistic method by building up a large scale, cross-process Bayesian Failure Network in lithium-ion battery production. Using this model, we are able to conduct root cause analyses as well as analyses of failure propagation. The former support operators in identifying root causes once a cell possesses a specific failure by calculating most-probable explanations matched to the individual battery cell data. The latter enable us to analyze propagation of failures and deviations in the production chain and thus provide support for placement of quality gates, leading to a significant reduction in scrap rate. Moreover, it gives an insight into which process steps are key drivers for which final product characteristics.

With the development of satellite load technology and very-large-scale integrated (VLSI) circuit technology, on-board real-time synthetic aperture radar (SAR) imaging systems have facilitated rapid response to disasters. Limited by severe size, weight, and power consumption constraints, a key challenge of on-board SAR imaging system design is to achieve high real-time processing performance. In addition, with the rise of multi-mode SAR applications, the reconfiguration of the on-board processing system is beginning to receive widespread attention. This paper presents a multi-mode SAR imaging chip with SoC architecture based on the reconfigurable double-operation engines and multilayer switching network. We decompose the commonly used extend chirp scaling (CS) SAR imaging algorithm into 8 types of double-operation engines according to the computing orders, and design a three-level switching network to connect these engines for data transition. The CPU is responsible for engine scheduling based on data flow driven with instructions to implement each part of the CS algorithm. Thus, multi-mode floating-point SAR imaging processing can be integrated into a single Application-Specific Integrated Circuit (ASIC) chip instead of relying on distributed technologies. As a proof of concept, a prototype measurement system with chip-included board is implemented, and the performance of the proposed design is demonstrated on Chinese Gaofen-3 stripmap continuous imaging. A chip requires 9.2 s, 50.6 s and 7.4 s for a stripmap with 16,384×16,384 granularity, multi-channel stripmap with 65.536×8192 granularity and multi-channel scan mode with 32,768×4096 granularity and 6.9 W for the system hardware to process the SAR raw data.

In daily travel and activities, pathfinding is a significant process. They are often used in transportation routes calculation. They have now evolved to be able to solve most situations of the pathfinding and its related problems. This review describes previous and recent studies on the pathfinding algorithms. It reviews the development of pathfinding algorithms in a classification base on their usage. The aim is to summarize the application of the pathfinding algorithms for the readers interested in the subject that can be used as a supplement.

The detection performance of small objects in remote sensing images is not satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) shows remarkable image enhancement performance, but reconstructed images miss high-frequency edge information. Therefore, object detection performance degrades for small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we apply a new edge-enhanced super-resolution GAN (EESRGAN) to improve the image quality of remote sensing images and use different detector networks in an end-to-end manner where detector loss is backpropagated into the EESRGAN to improve the detection performance. We propose an architecture with three components: ESRGAN, Edge Enhancement Network (EEN), and Detection network. We use residual-in-residual dense blocks (RRDB) for both the ESRGAN and EEN, and for the detector network, we use the faster region-based convolutional network (FRCNN) (two-stage detector) and single-shot multi-box detector (SSD) (one stage detector). Extensive experiments on a public (car overhead with context) and a self-assembled (oil and gas storage tank) satellite dataset show superior performance of our method compared to the standalone state-of-the-art object detectors.

Multi-Faith Spaces (MFS) are a relatively recent invention that quickly gained in significance. On the one hand, they offer a convenient solution for satisfying needs of people with diverse beliefs in the institutional context of hospitals, schools, airports, etc. On the other hand, as Andrew Crompton pointed out, they are politically significant because the multi-faith paradigm “is replacing Christianity as the face of public religion in Europe” (2012, p. 493). Due to their ideological entanglement, MFS are often used as the means to promote either a more privatised version of religion, or a certain denominational preference. Two distinct designs are used to achieve these means: negative in the case of the former, and positive in the latter. Neither is without problems, and neither adequately fulfils its primary purpose of serving diverse groups of believers. Both, however, seem to follow the biases and main problems of secularism. In this paper, I analyse recent developments of MFS to detail their main problems and answer the question, whether the MFS, and the underlying Multi-Faith Paradigm, can be classified as a continuation of secularism.

Citizen participation has always been of vital importance in decision-making processes and, likewise, joint work in the formulation and execution of projects that respond to territorial needs. This in order to ensure the best impact of public or private investment, optimizing the result and community articulation. For this reason, this investigation proposes a methodology of evaluation and prioritization of projects based on an approach integrated between the modelling in dynamics systems (SDM) and the Analytical process network (ANP), in which the citizen factors are used as qualitative and quantitative variables inside the posed balance sheet model. This is how the model developed by the Stella Architect software, it offers to the decision makers approximate information about the reality of the community, especially those who were affected by the Colombian conflict and where the expectations of a comprehensive repair could disrupt the socioeconomic system and reveal more pressing interests than those of the central government. The methodology is described by the information gotten through the participatory workshops done in the California county that belongs to the Magdalena region, this is a very affected region by the conflict in Colombia. The investigation results confirm the useful of, first of all, the use of the modelling as an information-generating tool by allowing the simulation of relevant variables in the evaluation and prioritization of projects at different moments of time and, in second place, use the knowledge of the experts that expose the ANP method for establishing the importance of each variable compared with other variables and, likewise of each project compared to the rest of the alternatives in each of the different instants of time in which the evaluation is carried out, given that the results could change in any time, the official ranking is showed after an optimization process with partial results.

Multi-channel and multi-parameter near-infrared spectrometers (NIRS) have gradually become a new research direction and hot spot due to their ability to provide real-time, continuous, comprehensive indicators of multiple parameters. However, multi-channel and multi-parameter detection may lead to crosstalk between signals. There is still a lack of benchmarks on the evaluation of reliability, sensitivity, stability and response consistency of the NIRS instrument. In this study, a set of test methods (the ink drop test, multi-channel crosstalk test and multi-parameter crosstalk test) for analysis crosstalk and verifying the reliability of NIRS were conducted to test experimental verification on a multi-channel (8-channel), multi-parameter (4-parameter) NIRS instrument independently developed by our team. Results show that these tests can be used to analyze the signal crosstalk and verify the reliability, sensitivity, stability and response consistency in the NIRS instrument. This study contributes to the establishment of benchmarks for the NIRS instrument crosstalk and reliability testing. These novel tests have the potential to become the benchmark for the NIRS instrument reliability testing.

Although there are several access control systems in the literature for flexible policy management in multi-authority and multi-domain environments, achieving interoperability & scalability, without relying on strong trust assumptions, is still an open challenge. We present HMBAC, a distributed fine-grained access control model for shared and dynamic multi-authority and multi-domain environments, along with Janus, a practical system for HMBAC policy enforcement. The proposed HMBAC model supports: (a) dynamic trust management between different authorities; (b) flexible access control policy enforcement, defined at domain and cross-domain level; (c) a global source of truth for all entities, supported by an immutable, audit-friendly mechanism. Janus implements the HMBAC model and relies on the effective fusion of two core components. First, a Hierarchical Multi-Blockchain architecture that acts as a single access point that cannot be bypassed by users or authorities. Second, a Multi-Authority Attribute Based Encryption protocol that supports flexible shared multi-owner encryption, where attribute keys from different authorities are combined to decrypt data distributedly stored in different authorities. Our approach was implemented using Hyperledger Fabric as the underlying blockchain, with the system components placed in Kubernetes Docker container pods. We experimentally validated the effectiveness and efficiency of Janus, while fully reproducible artifacts of both our implementation and our measurements are provided.

To define the optimistic multiple-granulation (OMG) supra interior operator and the OMG-supra closure operator associated with a set A in a multiple-granulation approximation space, we used the concepts of the OMG-lower set OL (A) and the OMG-upper set OU (A). These operators create an approximate topological space known as optimistic multiple-granulation supratopological spaces. OMG-supra-open and OMG-supra-closed approximation sets are also defined, followed by OMG-supra-continuous approximation and OMG-irresolute approximation.

The paper proposes TOPSIS method for solving multi-objective multi-level programming problem (MO-MLPP) with fuzzy parameters via fuzzy goal programming (FGP). At first, - cut method is used to transform the fuzzily described MO-MLPP into deterministic MO-MLPP. Then, for specific , we construct the membership functions of distance functions from positive ideal solution (PIS) and negative ideal solution (NIS) of all level decision makers (DMs). Thereafter, FGP based multi-objective decision model is established for each level DM for obtaining individual optimal solution. A possible relaxation on decisions for all DMs is taken into account for satisfactory solution. Subsequently, two FGP models are developed and compromise optimal solutions are found by minimizing the sum of negative deviational variables. To recognize the better compromise optimal solution, the concept of distance functions is utilized. Finally, a novel algorithm for MO-MLPP involving fuzzy parameters is provided and an illustrative example is solved to verify the proposed procedure.

We develop an interactive likelihood (ILH) for sequential Monte-Carlo (SMC) methods for image-based multiple target tracking applications. The purpose of the ILH is to improve tracking accuracy by reducing the need for data association. In addition, we integrate a recently developed deep neural network for pedestrian detection along with the ILH with a multi-Bernoulli filter. We evaluate the performance of the multi-Bernoulli filter with the ILH and the pedestrian detector in a number of publicly available datasets (2003 PETS INMOVE, AFL, and TUD-Stadtmitte) using standard, well-known multi-target tracking metrics (OSPA and CLEAR MOT). In all datasets, the ILH term increases the tracking accuracy of the multi-Bernoulli filter.

Vehicle logo detection plays a crucial role in various computer vision applications, such as vehicle classification and detection. In this research, we propose an improved vehicle logo detection method leveraging the MAMBA structure. The MAMBA structure integrates multiple attention mechanisms and bidirectional feature aggregation to enhance the discriminative power of the detection model. Specifically, we introduce the Multi-Head Attention for Multi-Scale Feature Fusion (MHAMFF) module to capture multi-scale contextual information effectively. Moreover, we incorporate the Bidirectional Aggregation Mechanism (BAM) to facilitate information exchange between different layers of the detection network. Experimental results on a benchmark dataset (VLD-45 dataset) demonstrate that our proposed method outperforms baseline models in terms of both detection accuracy and efficiency. Furthermore, extensive ablation studies validate the effectiveness of each component in the MAMBA structure. Overall, the proposed MAMBA-based vehicle logo detection approach shows promising potential for real-world applications in intelligent transportation systems.

Large-scale crowd phenomena are complex to model because the behaviour of pedestrians needs to be described at both strategic, tactical, and operational levels and is impacted by the density of the crowd. Microscopic models manage to mimic the dynamics at low densities, whereas mesoscopic models achieve better performances in dense situations. This paper proposes and evaluates a novel agent-based architecture to enable agents to dynamically change their operational model based on local density. The ability to combine microscopic and mesoscopic models for multi-scale simulation is studied through a use case of pedestrians at the Festival of Lights, Lyon, France. Pedestrian outflow data is extracted from video recordings of exiting crowds at the festival. The hybrid model is calibrated and validated using a genetic algorithm that optimizes the match between simulated and observed outflow data. Additionally, a local sensitivity analysis is then conducted to identify the most sensitive parameters in the model. Finally, the performance of the hybrid model is compared to different models in terms of density map and computation time. The results demonstrate that the hybrid model has the capacity to effectively simulate pedestrians across varied density scenarios while optimizing computational performance compared to other models.

One of the most important applications in the wireless sensor networks (WSN) is to classify mobile targets in the monitoring area. In this paper, a multi-DBN weighted voting classification algorithm is proposed on the basis of the Deep Belief Network (DBN) classifier and combined with the idea of voting method, which is implemented on the nodes of the WSN monitoring system by means of "upper training, lower transplantation" appraoch. The performance of the algorithm is verified by using real-world experimental data, and the results show that the proposed method has a higher accuracy in classifying the target signal features, achieving an average classification accuracy of 84.63% across four different types of moving targets. The experiment reveals that the multi-DBN weighted voting algorithm enhances the target classification accuracy by approximately 5% in comparison to the single DBN classifier, but the memory and computation time required for the algorithm to run are also increased at the same time. Compared to the FFNN classifier, which exhibited the highest classification accuracy among the four selected methods, the algorithm achieves an improvement of approximately 8.8% in classification accuracy. However, it incurs greater time overhead to run.

Unmanned Aerial Vehicles (UAVs) image object detection has great application value in military and civilian fields. However, the objects in the captured images from UAVs have problems of large scale variation, complex backgrounds, and a large proportion of small objects. To resolve these problems, a multi-scale object detector based on coordinate and global information aggregation is proposed, named CGMDet. Firstly, a Coordinate and Global Information Aggregation Module (CGAM) is designed by aggregating local, coordinate, and global information, which can obtain features with richer context information. Secondly, a Multi-Feature Fusion Pyramid Network (MF-FPN) is proposed, which can better fuse features of different scales and obtain features containing more context information through repeated use of feature maps, to better detect multi-scale targets. Moreover, more location information of low-level feature maps is integrated to improve the detection results of small targets. Furthermore, we modified the bounding box regression loss of the model to make the model more accurately regress the bounding box and faster convergence. Finally, the proposed CGMDet was tested on VisDrone and UAVDT datasets and mAP0.5 of 50.9% and 48% was obtained, respectively. At the same time, our detector achieved the best results compared to other detectors.

Rapid assessment of maize yields in smallholder farming system is important to understand its spatial and temporal variability and for timely agronomic decision-support. Imageries acquired with unmanned air vehicles (UAV) offer opportunity to assess agronomic variables at field scale, however, it is not clear if this can be translated into reliable yield assessment on smallholder farms where field conditions, maize genotypes, and management practices vary within short distances. In this study, we assessed the predictability of maize grain yield using UAV-derived vegetation indices (VI), with(out) biophysical variables, in smallholder farms. High-resolution images were acquired with UAV-borne multispectral sensor at 4 and 8 weeks after sowing (WAS) on 31 farmers’ managed fields (FMFs) and 12 nearby Nutrient Omission Trials (NOT), all distributed across 5 locations within the core maize region of Nigeria. The NOTs included non-fertilized and fertilized plots (with and without micronutrients), sown with open-pollinated or hybrid maize genotypes. Acquired multispectral images were post-processed into several three (s) vegetation indices (VIs), normalized difference vegetation index (NDVI), normalized difference red-edge (NDRE), green-normalized difference vegetation index (GNDVI). Biophysical variables, plant height (Ht) and percent canopy cover (CC), were measured with the georeferenced plot locations recorded. In the NOTs, the nutrient status, not genotype, influenced the grain yield variability and outcome. The maximum grain yield observed in NOTs was 9.3 tha^-1, compared to 5.4 tha^-1 in FMF. Without accounting for between- and within-field variations, there was no relationship between UAV-derived VIs and grain yield at 4WAS (r<0.02, P>0.1), but significant correlations were observed at 8WAS (r≤0.3; p<0.001). Ht was positively correlated with grain yield at 4WAS (r=0.5, R²=0.25, p<0.001), and more strongly at 8WAS (r=0.7, R²=0.55, p<0.001), while relationship between CC and yield was only significant at 8WAS. By accounting for within- and between-field variations in NOTs and FMF (separately) through linear mixed-effects modeling, predictability of grain yield from UAV-derived VIs was generally (R²≤0.24), however, the inclusion of ground-measured biophysical variable (mainly Ht) improved the explained yield variability (R² ≥0.62, RMSEP≤0.35) in NOTs but not in FMF. We conclude that yield prediction with UAV-acquired imageries (before harvest) is more reliable under controlled experimental conditions (NOTs), than in actual farmer-managed fields where various confounding agronomic factors can amplify noise-signal within the vegetation canopy.

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.

This project investigates the behavior of multi-head attention in Transformer models, specifically focusing on the differences between benign and trojan models in the context of sentiment analysis. Trojan attacks cause models to perform normally on clean inputs but exhibit misclassifications when presented with inputs containing predefined triggers. We characterize attention head functions in trojan and benign models, identifying specific 'trojan' heads and analyzing their behavior.

Urgent environmental challenges and emerging additive manufacturing (AM) technologies push research towards more performant and new materials. In the field of metallurgy, high entropy alloys (HEAs) have recently represented a topic of intense research because of their promising properties, such as high temperature strength and stability. Moreover, this class of multi-principal element alloys (MPEAs) have opened up researcher community to unexplored compositional spaces, making prosper literature of high-throughput methodologies and tools for rapidly screening large number of alloys. However, none of the methods has been aimed to design new MPEAs for AM process known as selective laser melting (SLM) so far. Here we conducted nanoindentation testing on single scan tracks of elemental powder blends and pre-alloyed powders after ball milling of AlTiCuNb and AlTiVNb. Results show that nanoindentation can represent an effective technique to gain information about phase evolution during laser scanning, contributing to accelerate the development of new MPEAs.

COP21 represented a starting point for several nations to develop and implement energy transition strategies to face and mitigate climate change, making the electrical power sector crucial in achieving the established goals and commitments. This research presents a 7-step methodology to improve decision-making in energy transition strategies for the power sector by integrating an economic dispatch optimization model based on linear programming to determine vulnerable aspects of power generation and transmission in strategic planning scenarios that could jeopardize the country’s energy transition. The application of the methodology is implemented under the open-source platform Python Optimization Modelling (PYOMO) and illustrated through a case study of the Mexican Electrical Power System (SEN) during the year 2025. The case study shows that the reserve margin fluctuated due to the variable renewable energy installed despite having a vast installed capacity to supply the country’s total demand. In addition, the results showed that most of the transmission lines had a congestion frequency higher than 90% of their capacity during most of the year. Two regions were identified as the best options for reducing greenhouse gas emissions by installing new power plants. Finally, most technologies reflected an under-generation, suggesting a high dependence on some fuels to supply the Mexican demand. The model’s programming is freely available in GitHub.

Smart Internet of Vehicles (IoVs) combined with Artificial Intelligence (AI) will contribute to vehicle decision-making in the Intelligent Transportation System (ITS). Multi-Vehicle Pursuit games (MVP), a multi-vehicle cooperative ability to capture mobile targets, is becoming a hot research topic gradually. Although there are some achievements in the field of MVP in the open space environment, the urban area brings complicated road structures and restricted moving spaces as challenges to the resolution of MVP games. We define an Observation-constrained MVP (OMVP) problem in this paper and propose a Transformer-based Time and Team Reinforcement Learning scheme (T3OMVP) to address the problem. First, a new multi-vehicle pursuit model is constructed based on decentralized partially observed Markov decision processes (Dec-POMDP) to instantiate this problem. Second, by introducing and modifying the transformer-based observation sequence, QMIX is redefined to adapt to the complicated road structure, restricted moving spaces and constrained observations, so as to control vehicles to pursue the target combining the vehicle’s observations. Third, a multi-intersection urban environment is built to verify the proposed scheme. Extensive experimental results demonstrate that the proposed T3OMVP scheme achieves significant improvements relative to state-of-the-art QMIX approaches by 9.66%~106.25%. Code is available at https://github.com/pipihaiziguai/T3OMVP.

Abstract: Thermal desalination is yet a reliable technology in the treatment of brackish water and seawater; however, its demanding high energy requirements have lagged it compared to other non-thermal technologies such as reverse osmosis. This review provides an outline of the development and trends of the three most commercially used thermal or phase change technologies worldwide: Multi Effect Distillation (MED), Multi Stage Flash (MSF), and Vapor Compression Distillation (VCD). First, state of water stress suffered by regions with little fresh water availability and existing desalination technologies that could become an alternative solution are shown. The most recent studies published for each commercial thermal technology are presented, focusing on optimizing the desalination process, improving efficiencies, and reducing energy demands. Then, an overview of the use of renewable energy and its potential for integration into both commercial and non-commercial desalination systems is shown. Finally, research trends and their orientation towards hybridization of technologies and use of renewable energies as a relevant alternative to the current problems of brackish water desalination are discussed. This reflective and updated review will help researchers to have a detailed state of the art of the subject and to have a starting point for their research, since current advances and trends on thermal desalination are shown.

The multi-UAV target search problem is crucial in the field of autonomous Unmanned Aerial Vehicle (UAV) decision-making. The algorithm design of Multi-agent Reinforcement Learning (MARL) methods has become integral to research on multi-UAV target search, due to its adaptability to the rapid online decision-making required by UAVs in complex, uncertain environments. In non-cooperative target search scenarios, targets may have the ability to escape, complicating UAVs' search efforts and hindering the convergence of the MARL algorithm training. This paper investigates the multi-UAV target search problem in scenarios involving static obstacles and dynamic escape targets, modeling the problem within the framework of Decentralized Partially Observable Markov Decision Process. Based on this model, a Spatio-Temporal Efficient Exploration network and a Global Convolutional Local Ascent mechanism are proposed. Subsequently,we introduce a Multi-UAV Escape Target Search Algorithm Based on MAPPO(ETS-MAPPO) for addressing the escape target search difficulty problem. Simulation results demonstrate that the ETS-MAPPO algorithm outperforms five classical MARL algorithms in terms of the number of target searches,the area coverage rate, and other metrics.