Search

The exponential growth of Internet of Things (IoT) devices underscores the need for robust security measures against cyber-attacks. Extensive research in the IoT security community has centered on effective traffic detection models, with a particular focus on anomaly intrusion detection systems (AIDS). This paper specifically addresses the preprocessing stage for IoT datasets and feature selection approaches to reduce the complexity of the data. The goal is to develop an efficient AIDS that strikes a balance between high accuracy and low detection time. To achieve this goal, we propose a hybrid feature selection approach that combines filter and wrapper methods. This approach is integrated into a two-level anomaly intrusion detection system. At level 1, our approach classifies network packets into normal or attack, with level 2 further classifying the attack to determine its specific category. One critical aspect we consider is the imbalance in these datasets, which is addressed using the Synthetic Minority Over-sampling Technique (SMOTE). To evaluate how the selected features affect the performance of the machine learning model across different algorithms, namely Decision Tree, Random Forest, Gaussian Naive Bayes, and k-Nearest Neighbor, we employ benchmark datasets: BoT-IoT, TON-IoT, and CIC-DDoS2019. Evaluation metrics encompass detection accuracy, precision, recall, and F1-score. Results indicate that the decision tree achieves high detection accuracy, ranging between 99.82 and 100%, with short detection times ranging between 0.02 and 0.15 s, outperforming existing AIDS architectures for IoT networks and establishing its superiority in achieving both accuracy and efficient detection times. [ABSTRACT FROM AUTHOR]

Identifying traffic signs is an essential part of traffic safety and self-driving systems. In real life, the driving environment is changing, making detecting traffic signs wisely and economically vital. The traffic sign detection problem has several small objects and complex ambient interference. The detecting situation also requires a practical and lightweight detection model. This study proposes a new lightweight model, the enhanced Ghost-YOLOv8, based on lightweight modules GhostConv and C3Ghost, based on the YOLOv8 model. It used a light method to extract the features, significantly speeding up inference. In addition to small, medium, and large objects, the head was expanded to include a new multi-scale feature extraction module layer focused on x-small. The experiment results show that when using the German Traffic Sign Detection Benchmark (GTSDB) dataset with three classes, the enhanced Ghost-YOLOv8 has mAP (0.50) of 99.4%and has fewer computations than the YOLOv8 model by 155.2 GFLOPs and has 18.6 Mparameters, which represents only 27.3% from the parameters used in the base model. Also, we suggested a new dataset called the GTSDB-43 dataset, which expanded the number of classes on the GTSDB dataset from three or four main classes to 43 classes and mentioned their main category type simultaneously. Compared with notable algorithms, this method's accuracy and speed are competitive. [ABSTRACT FROM AUTHOR]

There are currently many in-field methods for estimating soil properties (e.g., pH, texture, total C, total N) available in precision agriculture, but each have their own level of suitability and only a few can be used for direct determination of plant-available nutrients. As promising approaches for reliable in-field use, this review provides an overview of electromagnetic, conductivity-based, and electrochemical techniques for estimating plant-available soil nutrients and pH. Soil spectroscopy, conductivity, and ion-specific electrodes have received the most attention in proximal soil sensing as basic tools for precision agriculture during the last two decades. Spectral soil sensors provide indication of plant-available nutrients and pH, and electrochemical sensors provide highly accurate nitrate and pH measurements. This is currently the best way to accurately measure plant-available phosphorus and potassium, followed by spectral analysis. For economic and practicability reasons, the combination of multi-sensor in-field methods and soil data fusion has proven highly successful for assessing the status of plant-available nutrients in soil for precision agriculture. Simultaneous operation of sensors can cause problems for example because of mutual influences of different signals (electrical or mechanical). Data management systems provide relatively fast availability of information for evaluation of soil properties and their distribution in the field. For rapid and broad adoption of in-field soil analyses in farming practice, in addition to accuracy of fertilizer recommendations, certification as an official soil analysis method is indispensable. This would strongly increase acceptance of this innovative technology by farmers. [ABSTRACT FROM AUTHOR]

In the application of image super-resolution (SR) based on field-programmable gate array (FPGA), depthwise separable convolution is widely utilized. However, existing network designs overly simplify the structures used for deep feature extraction to conserve on-chip memory resources, which compromises network performance. Despite this, they still consume relatively large amounts of hardware resources. This study adopts a hardware-software co-design approach, proposing a hardware-friendly algorithm based on recurrent convolution neural network (RCNN), and implemented on FPGA. The algorithm employs a lightweight information multi-distillation block (LIMDB) to deeply extract and distill feature information, enhancing the network’s receptive field and improving its capability to extract feature information. Concurrently, the network reduces the dimensions of hidden state, enhancing performance and decreasing data transmission. Experimental results demonstrate that our lightweight deep recurrent convolution network (LDRCN) significantly outperforms other methods on common datasets. At the hardware level, we designed an efficient pipelining structure that combines skip connection and line buffer shared storage with a lossless segmentation computational strategy to reduce on-chip memory usage. Additionally, a method for finding mixed-precision quantization bit widths was designed to significantly reduce computational resource consumption while ensuring accuracy. The synthesis results on the ZYNQ XCZU9EG platform show that, compared to similar RNN-based works, the computation speed increased by 24.3% and DSP usage decreased by 43.5%, achieving ultra-high-definition (UHD) video output at 4K 95 fps. [ABSTRACT FROM AUTHOR]

Semantic segmentation is crucial in autonomous driving because of its accurate identification and segmentation of objects and regions. However, there is a conflict between segmentation accuracy and real-time performance on embedded devices. We propose an efficient lightweight semantic segmentation network (DRMNet) to solve these problems. Employing a streamlined bilateral structure, the model encodes semantic and spatial paths, cross-fusing features during encoding, and incorporates unique skip connections to coordinate upsampling within the semantic pathway. We design a new self-calibrated aggregate pyramid pooling module (SAPPM) at the end of the semantic branch to capture more comprehensive multi-scale semantic information and balance its extraction and inference speed. Furthermore, we designed a new feature fusion module, which guides the fusion of detail features and semantic features through attention perception, alleviating the problem of semantic information quickly covering spatial detail information. Experimental results on the CityScapes, CamVid, and NightCity datasets demonstrate the effectiveness of DRMNet. On a 2080Ti GPU, DRMNet achieves 78.6% mIoU at 88.3 FPS on the CityScapes dataset, 78.9% mIoU at 149 FPS on the CamVid dataset, and 53.5% mIoU at 160.4 FPS on the NightCity dataset. These results highlight the model’s ability to balance accuracy and real-time performance better, making it suitable for embedded devices in autonomous driving applications. [ABSTRACT FROM AUTHOR]

The recognition of alphabet signs is pivotal for the educational, linguistic, and cognitive development of individuals. This paper delves into the application of Convolutional Neural Networks (CNNs) on the ArASL database, a compilation of Arabic alphabet sign language images. Several CNN architectures underwent testing for image classification, with the EfficientNet B7 model emerging as the most effective, achieving an outstanding test accuracy of 99.24%. Capitalizing on the success of this optimized model, we have developed a real-time application with a dual purpose: to serve as a user-friendly tool for learning Arabic sign language and to enhance communication for individuals who are deaf or hard of hearing. The application, designed for seamless integration into users' daily lives, facilitates an intuitive engagement with the Arabic sign language alphabet. Its real-time functionality empowers users to learn and communicate effortlessly, breaking down barriers and fostering inclusivity in communication. This research contributes not only to the field of computer vision and deep learning but also to the broader spectrum of inclusive education and accessibility. By harnessing cutting-edge technology, our application represents a significant step forward in leveraging artificial intelligence for the betterment of education. [ABSTRACT FROM AUTHOR]

As an important data source for monitoring the behavior and variations of the ionosphere, the accuracy of current real-time global ionospheric maps (RT-GIMs) in low-latitude regions and oceanic regions is usually poor, and the accuracy during geomagnetic storms is not ideal. Therefore, the ionospheric vertical total electron content (VTEC) short-term forecast results were integrated into the global ionospheric real-time modeling process to improve the accuracy of RT-GIMs. Firstly, the preliminary RT-GIMs were established by constructing a virtual grid and determining the number of ionospheric pierce points in the grid. Then, different strategies were used to determine the virtual VTEC observations and filled the preliminary RT-GIMs. Finally, the filled RT-GIMs were modeled using spherical harmonic expansion and generated the final RT-GIMs, XRTG. On this basis, three ways were selected to evaluate the accuracy of XRTG. The GPS dSTEC (differential slant total electron content) assessment results showed that the performance of XRTG was the closest to that of Centre for Orbit Determination in Europe’s final GIMs (CODG), and it outperformed other RT-GIMs during geomagnetic storm periods and low-latitude regions. Compared with Universitat Politècnica de Catalunya’s RT-GIMs (UADG) with better performance in other RT-GIMs, the maximum decrease in root mean square error (RMSE) of XRTG during the geomagnetic storm period exceeds 25%, and the maximum decrease in the overall average RMSE of the 20 stations in low latitudes exceeds 27%. The Jason-3 VTEC assessment results showed that the accuracy of XRTG was closer to that of UADG and CODG, and the performance of XRTG and UADG in the range of 22° N–22° S was significantly better than that of other RT-GIMs. The consistency between XRTG and Universitat Politècnica de Catalunya’s rapid GIMs, Chinese Academy of Sciences’ final GIMs, and CODG was good, and the VTEC deviations from each post-processing GIMs were mainly concentrated in the range of ± 5 TECU. [ABSTRACT FROM AUTHOR]

Numerical simulations are crucial for predicting outcomes in forging processes but often neglect dynamic interactions within forming tools and presses. This study proposes an approach for achieving accurate real-time prediction of forging outcomes. Initially, a simulation-based surrogate model is developed to replicate key process characteristics related to the billet, enabling prediction of geometry, deformation field, and forging load after an upsetting operation. Subsequently, this model is integrated with a mass-spring-damper model representing the behavior of forging machine and tools. This integration enables the prediction of blow efficiency and energy distribution after each blow, including plastic, elastic, damping, and frictional energy of the upsetting operation. The approach is validated by comparing predictions with experimental results. The coupled model outperformed Finite Element Method (FEM) predictions, exhibiting mean absolute errors (MAE) below 0.1 mm and mean absolute percentage errors (MAPE) below 1% in geometry predictions. Deformation field predictions showed errors below 0.05 mm/mm, and load-displacement curves closely matched experimental data. Blow efficiency predictions aligned well with experimental results, demonstrating a mean absolute error below 1.1%. The observed energy distribution correlated with literature findings, underscoring the model’s fidelity. The proposed methodology presents a promising approach for accurate real-time prediction of forging outcomes. [ABSTRACT FROM AUTHOR]

Multicast has emerged as a primary communication pattern in datacenter networks due to the increasing demand for distributed data-parallel applications. To accelerate multicast traffic, the emerging reconfigurable circuit technology, which can establish circuit connections among switches, has been proposed as a promising paradigm for datacenter networks. This paper investigates how to accelerate the non-preemptive multicast flows in a demand-aware manner in reconfigurable datacenter networks. Firstly, the problem of scheduling circuit switches to minimize the average completion time is formulated and proved to be NP-hard. To address the conflicts between different multicast flows under the bandwidth constraint, a connection based hypergraph is constructed and then a two round matching algorithm is proposed under the bandwidth constraint. Additionally, to further reduce the average completion time, we introduce a method to utilize the remaining capacity of the ToR switches by splitting the unscheduled flows. The proposed algorithm is proved to have an approximation ratio of 2 2 n , where n represents the number of Top-of-Rack (ToR) switches. Finally, the extensive simulations demonstrate the effectiveness of the proposed algorithm in reducing the average completion time of flows compared to state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

The pressure decaying test is traditionally used to detect leakage in pneumatic actuators; however, it is time-consuming as it is performed between cycles of actuation. To date, there is a notable lack of real-time, data-driven machine learning methods to detect such leakages. Such leakage prediction is important to determine the lifecycle of the actuator and the exerting force to seal the vacuum chamber. This study aims to employ machine learning techniques to predict the categories of normal and leaked pressure loss ratio (PLR) in real-time during different number of pneumatic actuators cycles. The data was collected using a two-way factorial design with center points for the cycle test setting parameters. Seven parameters were recorded and undergo the training and test phase. Additionally, multiple model architecture was employed, each performing binary classification for each class to predict the PLR. Correlation analysis shows that pressure reading of each "open" and "close" inlets have better correlation for predicting the leakage with 0.216 and 0.183 respectively. The top performing model for each class is K-Nearest Neighbors (KNN) and Multilayer Perceptron (MLP). For the model 1, the KNN model achieved the highest accuracy of 93.86%, with an AUC of 0.94 in ROC analysis. Meanwhile, for the Model 2, utilizing MLP model attained an accuracy of 89.36% and an 84.87% mean-harmonic score. This study demonstrates that real-time machine learning can effectively predict leakage in pneumatic actuators, with the potential for further improvement through the inclusion of more data and increased setting points in the cycle test. [ABSTRACT FROM AUTHOR]

A new architectural approach to real-time embedded programming is described. A bare-metal program is written in C/C++ and combines event-driven technique with concurrency based on co-routines. Events are processed by soft real-time core at the priority level of software generated interrupts. Event is first posted to input queue of the core and then processed by invocation of its event handler. A special case of event is co-routine, its resumable function being a co-routine event handler. The co-routine that either yielded or needs to be resumed is queued for event processing once again. As a result, it is processed multiple times until execution of resumable function comes to the end of its operator sequence. Different levels of processing priority may be assigned to an event. Soft real-time core could be further expanded to run on symmetrical multiprocessor hardware. A combination of co-routines and basic events could easily be used in fork/join model. Concurrency constructs resemble those of Go and occam languages. Virtually all classic types of synchronization primitives could be implemented. The new approach was implemented for various ARM and Blackfin processors in C++ language as portable DORSECC library. This library was further used to program real-time embedded systems for mass-produced banknote sorting machines. One type of systems was used to recognize and validate banknote images by the method of cascade of one-class classifiers. The other system worked as a motion controller and used finite automata to control sensors and actuators. The total number of systems in operation is currently over 20000. The event and co-routine core in these systems provides average event processing time in the range of dozens of microseconds with sub-microsecond overhead time per each event. [ABSTRACT FROM AUTHOR]

Integrating the Internet of Things (IoT) with Social Networks (SN) has given rise to a new paradigm called Social IoT, which allows users and objects to establish social relationships. Nonetheless, trust issues such as attacks have emerged. These attacks can influence service discovery results. A trust management mechanism has become a major challenge in the Social IoT to prevent these attacks and ensure qualified services. A few studies have addressed trust management issues, especially those that prevent trust attacks in Social IoT environments. However, most studies have been dedicated to detect offline attacks with or without specifying the type of attack performed. These works will not be able to prevent attacks by aborting transactions between users because their primary purpose is to detect an offline attack. In addition, they do not consider security properties. This research paper aims to provide a detailed survey on trust management mechanism to handle trust attacks in Social IoT. In this research paper, we compared the techniques and technologies whose common point is attack prevention and demonstrated that blockchain technology can play a key role in developing a trust management mechanism that can prevent trust attacks while maintaining security properties. Then, we proposed combining the Apache Spark Framework with blockchain technology to provide real-time attack prevention. This combination can assist in creating upgraded trust management mechanisms in Social IoT environments. These mechanisms aim to prevent attacks in real-time through considering the security properties. Lack of survey papers in the area of trust attack prevention in real-time stands for an important motivational factor for writing this paper. The current research paper highlights the potential of the blockchain technology and Apache Spark in terms of developing an upgraded trust management able to prevent trust attacks in real-time.This paper provides a comprehensive survey on trust management mechanisms and approaches to handle trust attacks in Social IoT. Lack of such papers increases the significance of this paper. It also offers potential future research directions in terms of real-time trust attack prevention. [ABSTRACT FROM AUTHOR]

Symbolic generation of multibody systems equations of motion appeared in the 1980s. In addition to their computational advantage over their numerical counterparts, symbolic models can be very easily and straightforwardly interfaced with a wide range of software environments and hardware devices. These two features place this approach in a pole position to participate and intervene in the design of digital twins for systems such as vehicles, manipulators, walking robots or haptic devices. In this context, the first goal of this paper is to highlight the interest of symbolically generated multibody models – at the root of the ROBOTRAN program – in the form of a standalone set of equations calculating the dynamic model of multibody systems, for use as a computational component within a Digital-Twin-type process. The next goal is to embed realistic and complex multibody models within processes or devices whose functioning requires a synchronized real-time computation – or analysis – of their motion. An implementation (i) on specific hardware and (ii) on two extremely opposite but revealing applications (namely a railway vehicle and a digital piano) are presented to highlight the usefulness of symbolic models for the development of current and future multibody-based digital twins. [ABSTRACT FROM AUTHOR]

Purpose: Clinical needle insertion into tissue, commonly assisted by 2D ultrasound imaging for real-time navigation, faces the challenge of precise needle and probe alignment to reduce out-of-plane movement. Recent studies investigate 3D ultrasound imaging together with deep learning to overcome this problem, focusing on acquiring high-resolution images to create optimal conditions for needle tip detection. However, high-resolution also requires a lot of time for image acquisition and processing, which limits the real-time capability. Therefore, we aim to maximize the US volume rate with the trade-off of low image resolution. We propose a deep learning approach to directly extract the 3D needle tip position from sparsely sampled US volumes. Methods: We design an experimental setup with a robot inserting a needle into water and chicken liver tissue. In contrast to manual annotation, we assess the needle tip position from the known robot pose. During insertion, we acquire a large data set of low-resolution volumes using a 16 × 16 element matrix transducer with a volume rate of 4 Hz. We compare the performance of our deep learning approach with conventional needle segmentation. Results: Our experiments in water and liver show that deep learning outperforms the conventional approach while achieving sub-millimeter accuracy. We achieve mean position errors of 0.54 mm in water and 1.54 mm in liver for deep learning. Conclusion: Our study underlines the strengths of deep learning to predict the 3D needle positions from low-resolution ultrasound volumes. This is an important milestone for real-time needle navigation, simplifying the alignment of needle and ultrasound probe and enabling a 3D motion analysis. [ABSTRACT FROM AUTHOR]

Real-time time transfer is the basis for the time synchronization and establishment and maintenance of time scales. In this contribution, we present a regional multi-station real-time time transfer approach, in which observations from multi-global navigation satellite system (GNSS) received at regional multi-station are integrated to conduct an undifferenced network solution. Consequently, the simultaneous estimation of receiver clock offsets for multiple stations becomes possible, enabling the execution of real-time time transfer. One week of observation data from five multi-GNSS Experiment (MGEX) stations located in Europe are selected to generate four links (distances from 919.7 to 2153.4 km), and four schemes are designed, i.e., GPS-only, BDS-3-only, Galileo-only, and GPS/BDS-3/Galileo (GCE) solutions, respectively. Experiment results show that the mean number of observations, time dilution of precision (TDOP) values, and standard deviation (STD) of clock offset difference time series between the Center for Orbit Determination in Europe (CODE) and the estimated solutions for four time links are 90, 76, 72, 239 and 0.34, 0.38, 0.39, 0.21 and 0.039, 0.050, 0.043, 0.036 ns for GPS-only, BDS-3-only, Galileo-only, and GCE solutions, respectively. When the averaging time is shorter than 7680 s, the frequency stability of GCE solutions shows the best performance. The mean frequency stability of four time links at 7680 s is 6.59 × 10–15,7.25 × 10–15, 6.88 × 10–15 and 5.96 × 10–15 for GPS-only, BDS-3-only, Galileo-only, and GCE solutions, respectively. The GCE solution shows the best performance among the four schemes in terms of the number of observations, TDOP, STD, and frequency stability. The experiment results demonstrate that this approach is suitable for regional multi-station real-time time transfer. [ABSTRACT FROM AUTHOR]

Low Earth Orbit (LEO) satellites can augment the traditional GNSS-based positioning, navigation and timing services, which require real-time high-precision LEO satellite clock products. As the complicated systematic effects contained in the LEO satellite clock estimates limit their high-precision mid- to long-term prediction, high-frequency LEO satellite clocks need to be estimated within a Kalman filter, resulting in a short prediction time for real-time applications. Compared to the clock estimation using Batch Least-Squares (BLS) adjustment, filter-based clock estimation experiences a lower precision. Increasing the model strength by introducing external orbital information, thus, de-correlating the orbital and clock parameters, will benefit real-time clock precision. In this contribution, reduced-dynamic LEO satellite orbits are first estimated using BLS adjustment in near real-time and predicted in the short term. The predicted orbits are then constrained during the Kalman-filter-based clock estimation process. The variance–covariance matrix of the introduced orbital errors is tested for different sets of values in the radial, along-track and cross-track directions when constraining orbits of different prediction times. One week of GPS data from the Sentinel-3B satellite in 2018 was used for validation of the proposed method. When weakly constraining high-accuracy predicted orbits within a prediction time of 20 min, i.e., with a standard deviation of the constraint set to 2–3 dm in the radial and cross-track directions, and 4–6 dm in the along-track direction, the estimated clock accuracy can be improved from about 0.27 to 0.23 ns, with a 13.4% improvement. Depending on the prediction period of the introduced orbits, the Signal-In-Space Range Error (SISRE) of the LEO satellite to Earth can also be improved, from about 9.59 cm without constraints, to 7.38–8.07 cm after constraining the predicted orbits, with an improvement of 16–23%. The improvements in the SISRE also indicate a better consistency between the real-time clock and orbital estimates. [ABSTRACT FROM AUTHOR]

The tropospheric delay is difficult to be modeled accurately resulting from the high variability of atmospheric water vapor, especially under the conditions of sparse station distribution and large elevation differences, which poses challenges for real-time precise positioning. In this contribution, a real-time high-resolution (0.01° × 0.01°) zenith tropospheric delay (ZTD) model considering sparse stations and topography variations (named GFNSS) is established by integrating the information from the Global Forecast System (GFS) and Global Navigation Satellite System (GNSS). GNSS observations and GFS forecasts in the Hong Kong area are selected for the experiments. The performance of ZTDs derived from GFNSS is assessed and validated with the real-time GNSS ZTDs obtained by the precise point positioning method and the IGS post-processed ZTD products. Results show that the root mean square error (RMSE) of GFNSS ZTDs is 5.5 mm and 12.8 mm when validated with real-time and post-processed ZTD, while those for ZTD derived from the low-order surface model (LSM) are 8.8 mm and 19.0 mm, presenting a reduction of 37.5% and 32.6%, respectively. The sensitivity of model performance to the number of modeling stations and elevation differences is also evaluated. The results reveal that the GFNSS model is resistant to station number and presents high accuracy and stability with the RMSE values varying between 4.0 and 6.0 mm as the modeling station number decreases from 13 to 4, while the RMSE for the LSM model increases dramatically from 4.0 to 27.4 mm. Meanwhile, the GFNSS model achieves an RMSE value of 5.8 mm when the elevation differences are over 300 m, indicating a notable 84.9% reduction compared to that of LSM (RMSE of 38.5 mm). [ABSTRACT FROM AUTHOR]

Automatic extraction of coal flow region of coal mine belt conveyor plays an important role in coal flow monitoring, and real-time control of belt speed through real-time accurate monitoring of coal flow, which realizes the purpose of energy saving and consumption reduction of belt conveyor. In this paper, a real-time semantic segmentation network with detail enhancement for pixel-level coal flow monitoring, called DENet, is proposed. First, to ensure the strong real-time performance of the network, a two-branch coding structure is used to extract the semantic information and spatial detail information. Second, to improve the feature representation of spatial detail information, we design the Parameter-free Attention-Guided Enhancement Module (PF-AGEM) and the detail enhancement module (DEM), which fully integrate the semantic information features in the semantic branch into the detail branch and further enhance the detail features. Third, we design the multi-scale channel attention (MSCA) module in the semantic branch to extract the semantic information features of small targets earlier in the high-resolution feature maps, which solves the problem that the semantic information features of small targets are easily lost in the low-resolution feature maps. Finally, we propose a selective feature fusion module (FFM) to better realize the fusion of semantic information and spatial detail information. Experimental results show that the proposed DENet achieves a mean intersection over union (mIoU) of 96.23% at 87.1 frames per second (FPS) on the Coal Flow Segmentation (CFS) dataset and 74.9% mIoU at 207 FPS on the Camvid dataset, which is competitive with the state-of-the-art real-time semantic segmentation models. [ABSTRACT FROM AUTHOR]

Solar energy plays an important role in renewable energy generation, with the advantages of low pollution, easy installation, and relatively easy access. However, photovoltaic (PV) modules are susceptible to cause localized shading from external factors such as leaves in the canopy, surrounding buildings, etc., which would affect power generation efficiency and even pose safety risks. Existing methods cannot perform well in real-time conditions. This paper proposes a real-time shading monitoring method for the PV module based on computer vision. The gamma transform and histogram matching were adopted to enhance key features and adjust the global gamut strength distribution in the image of the PV module; then the gray-level slicing method finished the segmentation to detect the shadow from the video. All processing can be realized in the real-time monitor camera and the detection results can be displayed on the HMI in PC with high efficiency and low cost. According to tests in the practical complex environment, the method can have enough detection performance and high real-time performance with an accuracy of 0.98, and the F0.5 and F2 values are 0.87 and 0.85, respectively. The metrics of the proposed method are higher than those of the existing Canny detection method, the Random Forest detection method, and the CNN detection method. In addition, the average time required by the proposed method to process a frame is 0.721 s. In addition, the average time required by the method to process an image frame is 0.721 s, which has good real-time performance. [ABSTRACT FROM AUTHOR]

In Industry 4.0, the focus of task scheduling is more on smart services such as robotic services under the paradigm of mobile edge computing, which are widely used to improve the efficiency of smart manufacturing. Existing scheduling research efforts on optimizing efficiency such as swarm algorithm, reinforcement learning, but with little real-time dynamic scheduling. Online learning and adaptation is a critical function when tackling real challenges in unpredictable and dynamically changing edge computing environments. Our goal is to propose algorithms that can meet the real-time dynamic task scheduling even reach the level of communication. We propose a Deep Reinforcement Learning (DRL)-based online real-time task scheduling method which has the new exploration and exploitation strategy using the improved sparrow search algorithm (ISSA). We replace the traditional strategy with a meta-heuristic algorithm, while introducing natural disturbances and modifying the population position update formula to converge to the optimal position. The algorithm has two layers. The first layer is scheduling generation layer, which is responsible for making decisions, and the second layer is scheduling policy update layer, which optimizes the strategy and finds the best set of hyperparameters. The experimental results show that the running time of the method to solve the problem can reach 0.0092 seconds, and the real-time performance reaches up to the millisecond communication level. At the same time other methods are compared, effectively reducing the running cost and meeting the requirements of production scheduling tasks. [ABSTRACT FROM AUTHOR]

In this work, a stochastic dynamic version of the container drayage problem is studied. The presented model incorporates uncertainty in the form of stochastic loading and unloading times at both terminals and customers, as well as stochastic travel times, conditionally dependent upon the departure time, allowing robust planning with respect to varying processing times. Moreover, the presented model is dynamic, allowing flexible orders and having the capability of re-solving the optimization problem in case of last-minute orders. Finally, the model also incorporates a truck appointment system operating at each terminal. First, a description of the general model is given, which amounts to a mixed integer non-linear program. In order to efficiently solve the optimization problem, and linearize both the objective and the conditional chance constraints, it is reformulated based on time window partitioning, yielding a purely integer linear program. As a test case, a large road carrier operating in the port of Antwerp is considered. We demonstrate that the model is efficiently solvable, even for instances of up to 300 orders. Moreover, the impact of incorporating stochastic information is clearly illustrated. [ABSTRACT FROM AUTHOR]

There is a growing need in public health to conduct large-scale epidemiological studies to investigate the health effects of fine particulate matter (PM2.5) exposure levels. In response to this need, we developed a real-time personal PM2.5 exposure monitoring system (PEPS: Personal Exposure PM2.5 System), which is capable of monitoring personal exposure concentration and uploading data in real time. The air quality self-labelling device, specifically customized for the PEPS, can be worn on the body and features functions for real-time data automatic upload, data storage, data export, and localization. This system enables researchers to obtain the big data of personal PM2.5 exposure concentration at low cost, with minimal manpower and technical requirements. It has been utilized to investigate the personal exposure levels of PM2.5 among college students in Beijing, China, providing a substantial volume of valuable data for indoor air quality and related epidemiological study. The maximum difference between the monitored daily average exposure concentration and the outdoor concentration was 265 µg/m3, corresponding to a relative error of 1579.5%. The correlation analysis of 11 factors showed that the correlation between exposure concentration and outdoor concentration was as high as 0.66 (p < 0.001), and the correlation between exposure concentration and other certain factors was in the range of [−0.11, −0.03]. [ABSTRACT FROM AUTHOR]

Intelligent transportation systems rely heavily on accurate traffic sign detection (TSD) to enhance road safety and traffic management. Various methods have been explored in the literature for this purpose, with deep learning methods consistently demonstrating superior accuracy. However, existing research highlights the persistent challenge of achieving high accuracy rates while maintaining non-destructive and real-time requirements. In this study, we propose a deep learning model based on the YOLOv8 architecture to address this challenge. The model is trained and evaluated using a custom dataset, and extensive experiments and performance analysis demonstrate its ability to achieve precise results, thus offering a promising solution to the current research challenge in deep learning-based TSD. [ABSTRACT FROM AUTHOR]

Meeting the rising global demand for healthcare diagnostic tools is crucial, especially with a shortage of medical professionals. This issue has increased interest in utilizing deep learning (DL) and telemedicine technologies. DL, a branch of artificial intelligence, has progressed due to advancements in digital technology and data availability and has proven to be effective in solving previously challenging learning problems. Convolutional neural networks (CNNs) show potential in image detection and recognition, particularly in healthcare applications. However, due to their resource-intensiveness, they surpass the capabilities of general-purpose CPUs. Therefore, hardware accelerators such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), and graphics processing units (GPUs) have been developed. With their parallelism efficiency and energy-saving capabilities, FPGAs have gained popularity for DL networks. This research aims to automate the classification of normal and abnormal (specifically Diabetic Foot Ulcer—DFU) classes using various parallel hardware accelerators. The study introduces two CNN models, namely DFU_FNet and DFU_TFNet. DFU_FNet is a simple model that extracts features used to train classifiers like SVM and KNN. On the other hand, DFU_TFNet is a deeper model that employs transfer learning to test hardware efficiency on both shallow and deep models. DFU_TFNet has outperformed AlexNet, VGG16, and GoogleNet benchmarks with an accuracy 99.81%, precision 99.38% and F1-Score 99.25%. In addition, the study evaluated two high-performance computing platforms, GPUs and FPGAs, for real-time system requirements. The comparison of processing time and power consumption revealed that while GPUs outpace FPGAs in processing speed, FPGAs exhibit significantly lower power consumption than GPUs. [ABSTRACT FROM AUTHOR]

With the growth of computer networks worldwide, there has been a greater need to protect local networks from malicious data that travel over the network. The increase in volume, speed, and variety of data requires a more robust, accurate intrusion detection system capable of analyzing a huge amount of data. This work proposes the creation of an intrusion detection system using stream classifiers and three classification layers—with and without a reduction in the number of features of the records and three classifiers in parallel with a voting system. The results obtained by the proposed system are compared against other models proposed in the literature, using two datasets to validate the proposed system. In all cases, gains in accuracy of up to 18.52% and 3.55% were obtained, using the datasets NSL-KDD and CICIDS2017, respectively. Reductions in classification time up to 35.51% and 94.90% were also obtained using the NSL-KDD and CICIDS2017 datasets, respectively. [ABSTRACT FROM AUTHOR]

In this paper a system to monitor road conditions, detect unsafe driving behaviours and determine the influence of rainfall on traffic safety in real time using different machine learning algorithms, has been proposed. The system developed consists of a mobile application that captures car movement using its in-built accelerometer and gyroscope sensors and a server that monitors weather conditions at 16 key locations in Mauritius using the OpenWeather API. Road conditions, pothole, speed bumps as well as driving events were analysed using the K-Nearest Neighbour (KNN) and Multi-Layer Perceptron (MLP) algorithms. Moreover, a mathematical model, which incorporates the predicted rainfall in the estimation of braking distance and recommended speed, has been proposed. An average accuracy of 80.9% was obtained for pothole detection, 70% for speed bumps and 85.5% for unsafe driving behaviours detection. The proposed model with rainfall data predicted the braking distance and recommended speed with a Mean Absolute Percentage Error (MAPE) of 14.7% and 0.735% respectively. [ABSTRACT FROM AUTHOR]

Despite sufficient studies performed in non-primate animal models, there exists scanty information obtained from pilot trials in non-human primate animal models, severely hindering nanomaterials moving from basic research into clinical practice. We herein present a pioneering demonstration of nanomaterials based optical imaging-guided surgical operation by using macaques as a typical kind of non-human primate-animal models. Typically, taking advantages of strong and stable fluorescence of the small-sized (diameter: ~ 5 nm) silicon-based nanoparticles (SiNPs), lymphatic drainage patterns can be vividly visualized in a real-time manner, and lymph nodes (LN) are able to be sensitively detected and precisely excised from small animal models (e.g., rats and rabbits) to non-human primate animal models (e.g., cynomolgus macaque (Macaca fascicularis) and rhesus macaque (Macaca mulatta)). Compared to clinically used invisible near-infrared (NIR) lymphatic tracers (i.e., indocyanine green (ICG); etc.), we fully indicate that the SiNPs feature unique advantages for naked-eye visible fluorescence-guided surgical operation in long-term manners. Thorough toxicological analysis in macaque models further provides confirming evidence of favorable biocompatibility of the SiNPs probes. We expect that our findings would facilitate the translation of nanomaterials from the laboratory to the clinic, especially in the field of cancer treatment. [ABSTRACT FROM AUTHOR]

Global fire performance of structures in fire is proven to be more advantageous in many cases of engineering practice than the prescriptive fire resistance based on isolated structural member testing. Hybrid fire simulation (HFS) is a novel well-suited method trending in recent years for analysis of global performance of structures in fire. In the principles of this method, the part of a structure which has unknown behavior or is uncertain to be numerically modeled (subjected to fire) would be physically tested, while the rest of the structure is numerically simulated. HFS method enables capturing the beneficial interaction mechanisms evolving between fire-exposed structural members and the adjacent cooler substructure. Due to the continuous temperature increase in a fire test and the existing thermal inertia as well as the rate- and temperature-dependent material behavior of structures exposed to fire, a real-time performance in hybrid fire simulation counts as a necessity. This challenge is more critical for hybrid fire simulations with higher applied heating rates relevant to structural fire engineering. Within scope of this paper, (a) a robust and rigorous approach for real-time HFS is presented; (b) a series of proof-of-concept studies of different hybrid fire simulations with various applied heating rates are carried out for a thermomechanical benchmark problem; (c) the important results of four representative hybrid fire simulations with relevant heating rates to structural fire engineering are discussed; (d) the importance of an appropriate calculation method for stiffness update of the fire-exposed structural member over HFS procedure is highlighted, and e) the precision and accuracy of the applied HFS approach with respect to interface error and real-time degree are evidenced. [ABSTRACT FROM AUTHOR]

The huge growth in the number of vehicles is causing serious traffic management problems. Existing roads must handle traffic more than expected which presents serious challenges including congestion and safety. Intelligent traffic management systems were proposed as a solution to solve such problems. This intelligent system is charged to estimate vehicle density and manage the traffic flow accordingly. A wide variety of sensors was deployed for the traffic management system such as cameras, ultrasonic sensors, radar, infrared sensors, and acoustic sensors. In this work, an intelligent traffic flow management system was proposed based on data provided by public surveillance cameras. For this purpose, a real-time vehicle detection and counting model based on You Look Only Once (YOLO) v6 was deployed. The proposed model was trained and evaluated on different publicly available datasets such as BDD 100K and KITTI. Extensive experimentations proved the efficacy of the proposed model for detecting vehicles while operating under real-time constraints. The proposed traffic flow management system was evaluated using real videos for different traffic scenarios, and good results were achieved. [ABSTRACT FROM AUTHOR]

Distributed hybrid rendering (DHR) is a real-time rendering approach that incorporates cloud-based ray tracing with locally rasterized graphics for interactive thin client metaverse and game applications. With cloud assistance, DHR can generate high-fidelity ray-traced graphics contents remotely and deliver them to thin clients with low graphics capability, including standalone extended reality devices and mobile phones, while maintaining interactive frame rates for users under adverse network conditions. DHR can already achieve the effect of ray-traced hard shadows that form with the occlusion of direct illumination. We enhance the realism of these shadows by softening their edges with the direction of rays traced and approximating the occlusion of indirect illumination by reconstructing ray-traced ambient occlusion with a modified version of spatiotemporal variance-guided filtering. Our technique uses only 20–30% of the bandwidth of remote rendering and is also tolerant of delays of up to 200 ms with only slight distortion to the shadows along object edges. [ABSTRACT FROM AUTHOR]

Nowadays, more and more brands use interesting anime characters to promote and increase brand awareness. However, real-time and interesting promotional materials are also one of the key factors to attract people, so real-time processing of anime characters has also become an effective way to enhance brand awareness. In recent years, with the rapid development of deep learning, image style conversion based on AI technology has become a much-attended artificial intelligence application, but it also suffers from the disadvantages of complex model structure, slow conversion speed, and inconspicuous identity features, which need to be improved. In view of this, this study proposes an anime portrait realization (ARF-GAN) algorithm based on generative adversarial network. This algorithm is based on the Pix2Pix architecture and also uses the U-net network structure with jump connections to connect the encoder’s feature maps directly to the decoder, which in turn enables the network to reconstruct the output data in a better way and the network architecture is more lightweight.This algorithm is based on the Pix2Pix architecture and also uses the U-net network structure with jump connections to connect the encoder’s feature maps directly to the decoder, which in turn enables the network to reconstruct the output data in a better way and the network architecture is more lightweight. It also introduces the CBAM module, which can enhance the model’s expressive and generative capabilities without increasing the model’s complexity, and improve the model’s real-time image processing capability. In addition, this paper also compensates for the problem of image blurring brought by the original architecture by introducing the deblurring module. The experimental results on CartoonFaceAB, DanBooru show that the proposed ARF-GAN has a better generative effect for the task of realizing anime portraits. And comparing with different types of image generating models, it shows better accuracy and lower complexity by using several evaluation indexes such as PSNR and SSIM. This makes it more suitable for brand advertisement promotion, so it has good application value for improving brand awareness. [ABSTRACT FROM AUTHOR]

Real-time precise orbit and clock products are prerequisites for Real-Time Precise Point Positioning (RT-PPP) and its related applications, such as time synchronization and disaster monitoring. While real-time products have achieved relatively high accuracy, occasional outliers and accuracy degradation significantly restricts the application of RT-PPP in safety-critical fields. Sub-meter anomalies or larger are typically manageable, as users can easily detect and exclude them through outlier detection in the preprocessing stage before positioning. However, handling small-scale anomalies solely with user-side data and algorithms poses challenges, which also affect the accuracy and reliability of positioning solutions. To address this, we propose a quality monitoring method for real-time precise satellite orbit and clock products. The method utilizes a quality monitoring network of well-distributed stations to validate real-time products continuously. The product quality monitoring server-side calculates both pseudorange and carrier-phase Quality Indicator (QI) for each satellite by using real-time statistics of residuals from quality monitoring network stations, supplemented by product error empirical models. Furthermore, the isolation Forest (iForest) algorithm is employed to detect outliers prior to real-time residual statistics, mitigating the impact of monitoring network or communication link failures on QI while slightly increasing the computational load. QI are broadcasted to users, aiding them in excluding satellites with lower accuracy or reducing the weighting of these satellites in the positioning solution. We conducted a 1 month quality monitoring of Centre National d'Etudes Spatiales real-time orbit and clock products in January 2023, using 30 continuous stations of the Crustal Movement Observation Network of China. The results indicate that the pseudorange QI is greater than 1 m, which can adequately bound the product errors but still exhibit considerable redundancy. In contrast, the carrier-phase QI is within 15 cm, significantly reducing redundancies. The carrier-phase QI can instantly and accurately reflect the accuracy changes of satellite orbit and clock products and bound more than 99.93% of the product errors. The addition of iForest outlier detection reduces the number of false alert epochs from 74 to 21, reducing the false alert probability by 68.9%. At the same time, the execution time per epoch slightly increased from 28.8 to 32.1 ms, representing an 11.0% increase. Importantly, this increase in execution time does not significantly impact the Time to Alert. In addition, QI can effectively detect step faults with amplitudes greater than 3 cm and ramp faults with slopes greater than 0.05 cm/s. QI shows an immediate increase when faults occur and quickly returns to normal after faults end. [ABSTRACT FROM AUTHOR]

Dense Global Navigation Satellite System (GNSS) observations are beneficial for monitoring the small-scale troposphere turbulence in both spatiotemporal domains. Densification of current geodetic GNSS networks can be achieved by incorporating low-cost GNSS receivers and antennas, benefitting weather monitoring. This study aims to focus on real-time tropospheric delay estimation using real-time GNSS data streams collected by low-cost GNSS receivers and antennas. The results indicate that: (1) Low-cost GNSS devices can offer accuracies of 7.5 mm in winter and 10.8 mm in summer for real-time zenith tropospheric delay (ZTD) estimations; (2) By applying the phase center variation corrections for low-cost GNSS antennas, which are often ignored in previous studies on low-cost GNSS applications, can produce more reliable real-time ZTD estimations; (3) Low-cost GNSS devices can capture the ZTD fluctuations under heavy precipitation events in a real-time mode, showing a great potential as sensors to monitoring severe weather. These results underscore the viability of low-cost GNSS devices as cost-effective sensors that complement geodetic GNSS equipment in real-time troposphere monitoring and forecasting, especially at small scales. Their integration aids in the densification of existing monitoring networks, thereby bolstering GNSS meteorology applications. [ABSTRACT FROM AUTHOR]

Stereo matching methods based on convolution neural network (CNN) often face challenges such as edge blurring and the loss of small structures. These issues often result in incorrect disparity assignments when upsampling the disparity map. To address this problem, we propose a disparity refinement module (GDU-CTF) that combines guided disparity map upsampling with a coarse-to-fine process. This approach effectively restores incorrect disparity values in the final disparity map. Furthermore, due to the insufficient aggregation of global geometric and contextual texture features using basic encoder–decoder 3D convolutional networks, we propose a guided patch cost aggregation module (GPA) that generates a more precise initial disparity map for textureless areas. These modules complement each other and are efficient, resulting in an accurate and lightweight framework for stereo matching. Experimental results demonstrate that our algorithm has excellent accuracy in generating disparity maps and achieves outstanding real-time performance, with an inference time of just 0.03 s on Scene Flow and KITTI datasets. [ABSTRACT FROM AUTHOR]

Copyright of Frontiers of Information Technology & Electronic Engineering is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In the Internet of Medical Things (IoMT)-based real-time telemedicine systems, patients can utilize a wide range of medical devices and sensors, which leads to the continuous generation of massive amounts of data. The high speed of data generation poses challenges in collecting, organizing, processing, and making decisions about patients' emergency levels. Existing methods for classifying (triaging) patients in such environments often yield inaccurate triage levels, necessitating a computational approach to enhance accuracy. This research aims to handle data from multiple heterogeneous sources in IoMT-based real-time telemedicine systems, analyze the data to accurately triage patients with the most urgent cases, and provide swift healthcare services. The proposed solution, the Data Processing with Triaging Model (DPTM), employs a hybrid approach that combines principal component analysis and decision tree algorithms. The model was tested using 55,680 patients with two chronic diseases: heart disease and hypertension. The computational results demonstrate the promising effectiveness of DPTM in accommodating and managing the requests of 55,680 patients. The proposed system achieves an impressive accuracy of 93%, surpassing four other algorithms. In conclusion, DPTM enhances medical services, reduces hospital overcrowding, and ensures accurate services for all patients, with a 93% accuracy rate. [ABSTRACT FROM AUTHOR]

Ransomware, particularly crypto ransomware, has emerged as the go-to malware for threat actors aiming to compromise data on Android devices as well as in general. In this paper, we present a ransomware detection technique based on behaviours observed in the system calls performed by the malware. We first describe our repeatable and extensible methodology for extracting the system call log and patterns. We then identify and present some common high-level system call behavioural patterns exhibited by crypto ransomware, and evaluate these patterns. We further describe the implementation of a streaming implementation that utilises regular expressions for modelling malware behaviours and finite state machines for detecting crypto ransomware behaviours in real time. The success of our proof of concept evaluation allows us to envision our proposed technique applied as part of a self-protection system on Android phones against malware. [ABSTRACT FROM AUTHOR]

Instance segmentation of foods is an important technology to ensure the food success rate of meal-assisting robotics. However, due to foods have strong intraclass variability, interclass similarity, and complex physical properties, which leads to more challenges in recognition, localization, and contour acquisition of foods. To address the above issues, this paper proposed a novel method for instance segmentation of foods. Specifically, in backbone network, deformable convolution was introduced to enhance the ability of YOLOv8 architecture to capture finer-grained spatial information, and efficient multiscale attention based on cross-spatial learning was introduced to improve sensitivity and expressiveness of multiscale inputs. In neck network, classical convolution and C2f modules were replaced by lightweight convolution GSConv and improved VoV-GSCSP aggregation module, respectively, to improve inference speed of models. We abbreviated it as the DEG-YOLOv8n-seg model. The proposed method was compared with baseline model and several state-of-the-art (SOTA) segmentation models on datasets, respectively. The results show that the DEG-YOLOv8n-seg model has higher accuracy, faster speed, and stronger robustness. Specifically, the DEG-YOLOv8n-seg model can achieve 84.6% Box_mAP@0.5 and 84.1% Mask_mAP@0.5 accuracy at 55.2 FPS and 11.1 GFLOPs. The importance of adopting data augmentation and the effectiveness of introducing deformable convolution, EMA, and VoV-GSCSP were verified by ablation experiments. Finally, the DEG-YOLOv8n-seg model was applied to experiments of food instance segmentation for meal-assisting robots. The results show that the DEG-YOLOv8n-seg can achieve better instance segmentation of foods. This work can promote the development of intelligent meal-assisting robotics technology and can provide theoretical foundations for other tasks of the computer vision field with some reference value. [ABSTRACT FROM AUTHOR]

The ever-increasing demand for high performance in the time-critical, low-power embedded domain drives the adoption of powerful but unpredictable, heterogeneous Systems-on-Chip. On these platforms, the main source of unpredictability—the shared memory subsystem—has been widely studied, and several approaches to mitigate undesired effects have been proposed over the years. Among them, performance-counter-based regulation methods have proved particularly successful. Unfortunately, such regulation methods require precise knowledge of each task's memory consumption and cannot be extended to isolate mixed-criticality tasks running on the same core as the regulation budget is shared. Moreover, the desirable combination of these methodologies with well-known time-isolation techniques—such as server-based reservations—is still an uncharted territory and lacks a precise characterization of possible benefits and limitations. Recognizing the importance of such consolidation for designing predictable real-time systems, we introduce MCTI (Mixed-Criticality Task-based Isolation) as a first initial step in this direction. MCTI is a hardware/software co-design architecture that aims to improve both CPU and memory isolations among tasks with different criticalities even when they share the same CPU. In order to ascertain the correct behavior and distill the benefits of MCTI, we implemented and tested the proposed prototype architecture on a widely available off-the-shelf platform. The evaluation of our prototype shows that (1) MCTI helps shield critical tasks from concurrent non-critical tasks sharing the same memory budget, with only a limited increase in response time being observed, and (2) critical tasks running under memory stress exhibit an average response time close to that achieved when running without memory stress. [ABSTRACT FROM AUTHOR]

Recent flood events (FE) in Germany have shown that the extent and impact of extreme flood events cannot be estimated solely based on numerical models. For analyzing the development of such an event and to develop and implement safety measures more efficiently, additional data must be collected during the event. Within the scope of this research, the possibilities of near real-time recording using an unmanned aerial vehicle (UAV) and data processing with the Structure from Motion (SfM) method were tested in a case study. Different recording parameter combinations were tested in the Laufer Muehle area on the Aisch river in Germany. The focus of the investigations was the identification of a parameter combination that allows a short recording interval for aerial imagery. Based on these findings, the identification of changes in the study area by comparing multitemporal photography (flood prevention), as well as the recording of flooded areas during a FE should be possible. The accuracy analysis of the different parameter combinations between two point clouds as well as the process of change detection was done by a Multiscale Model to Model Cloud Comparison (M3C2) and including ground control points. As a result, a parameter combination was identified which led to the desired results in the study area. The processes were transformed into fully automated and scripted workflows. The results serve as a basis for establishing a workflow for near real-time analyses in future studies. [ABSTRACT FROM AUTHOR]

The development of safety-critical real-time embedded systems (RTESs) is a difficult task. Indeed, in addition to functional requirements, these systems must guarantee timing constraints. Model-driven engineering (MDE) promotes a rise in the level of abstraction during the development process of these systems. This is beneficial since it may help designers manage the increasing complexity of these systems. It may also accelerate the development process through model transformations and enable early verification of the timing properties. Schedulability analysis techniques aim to verify whether the tasks complete their executions within the time limit specified by the real-time application (i.e., the deadline). Unfortunately, software designers find this task difficult since it requires a deep understanding of real-time scheduling theory, as well as the use of dedicated tools (called analysis tools). In this paper, we propose real-time verification design patterns to promote early verification of timing properties in a model-based approach. These patterns aim to assist designers in the verification process. They encapsulate the knowledge of a real-time expert in scheduling theory. To facilitate and fasten the verification process, we propose to automate the generation of the analysis model. Thanks to the provided patterns, this generation phase includes an automatic consistency check of the design model. The consistency check step verifies the design's validity and completeness with respect to the applied pattern. To add further benefits to the proposed patterns, we describe a process that provides context for how these patterns can be applied. We apply and simulate the contribution to real-world case study. The obtained results show the viability of this research. [ABSTRACT FROM AUTHOR]

In this brief, a real-time inversion of petrophysical information in downhole firmware for nuclear magnetic resonance logging while drilling (NMR-LWD) based on a field-programmable gate array (FPGA) is proposed. Firstly, the inverse problem is introduced, the nonnegative least-squares (NNLS) problem is modeled and solved by the accelerated projected gradient descent (APGD) algorithm. Secondly, the efficient architecture for the APGD accelerator based on look-up tables (LUTs) is used in Intel's 10M50 FPGA. Finally, a software cooling method based on a multi-core structure and periodic thermal management technology is deployed. The proposed scheme is applied to NMR-LWD, the effectiveness and feasibility of the proposed scheme are verified experimentally. [ABSTRACT FROM AUTHOR]

Objective: Clinical-standard MRI is the imaging modality of choice for the wrist, yet limited to static evaluation, thereby potentially missing dynamic instability patterns. We aimed to investigate the clinical benefit of (dynamic) real-time MRI, complemented by automatic analysis, in patients with complete or partial scapholunate ligament (SLL) tears. Material and methods: Both wrists of ten patients with unilateral SLL tears (six partial, four complete tears) as diagnosed by clinical-standard MRI were imaged during continuous active radioulnar motion using a 1.5-T MRI scanner in combination with a custom-made motion device. Following automatic segmentation of the wrist, the scapholunate and lunotriquetral joint widths were analyzed across the entire range of motion (ROM). Mixed-effects model analysis of variance (ANOVA) followed by Tukey's posthoc test and two-way ANOVA were used for statistical analysis. Results: With the increasing extent of SLL tear, the scapholunate joint widths in injured wrists were significantly larger over the entire ROM compared to those of the contralateral healthy wrists (p<0.001). Differences between partial and complete tears were most pronounced at 5°–15° ulnar abduction (p<0.001). Motion patterns and trajectories were altered. Complete SLL deficiency resulted in complex alterations of the lunotriquetral joint widths. Conclusion: Real-time MRI may improve the functional diagnosis of SLL insufficiency and aid therapeutic decision-making by revealing dynamic forms of dissociative instability within the proximal carpus. Static MRI best differentiates SLL-injured wrists at 5°–15° of ulnar abduction. [ABSTRACT FROM AUTHOR]

Thomson scattering of the core edge and divertor plasma regions of a tokamak with reactor technologies is discussed. The rationale and choice of technical solutions are given, the composition of the Thomson scattering diagnostic complex is discussed, as well as an estimate of the accuracy of measuring both electron temperature and density. Particular attention is paid to ensuring the functionality of the proposed diagnostics in the reactor mode of the tokamak operation and the results of testing diagnostic equipment in the experiments on Globus-M2 tokamak. [ABSTRACT FROM AUTHOR]

Extended Kalman filtering with unknown input (EKF-UI) is often used to estimate the structural system state, parameters and unknown input in structural health monitoring. However, the real-time performance of EKF-UI is bound to whether the measurement equation has a direct feedthrough of unknown input, which great limits its application scope. Based on the zero-order-hold assumption and random walk assumption of unknown input, a novel adaptive discrete state equation is derived in this paper. The new equation establishes a connection between the current state and the current input and allows the adjustment of the sensitivity matrix of the unknown input. Then, based on the adaptive discrete state equation and minimum variance unbiased estimation principle, an adaptive generalized extended Kalman filter with unknown input is derived. The proposed algorithm eliminates the limitation that the real-time performance is restricted by whether the measurement equation has a direct feedthrough of the input and realizes the optimization of the state and input estimates in the sense of minimum variance. To demonstrate the feasibility of the proposed method, numerical example of a shear frame structure with Bouc–Wen hysteresis nonlinearity and experimental test of a five-story shear frame are conducted. The comparison with existing methods shows the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, a marker and its real-time tracking method are proposed. Different from existing technologies that recognize the markers by spatial event features, which lead to many practical problems, we discover the possibility of using temporal features. Strobe LEDs (light emitting diode) are used as markers to produce periodically flipping events, and a fast clustering-based algorithm is designed to track and recognize these markers simultaneously. Experiments demonstrate that our methods have superior speed and accuracy compared to state-of-the-arts. The markers can be stably tracked in many challenging situations, thus can be used in various visual tracking applications. The proposed method introduces a new marker and its corresponding recognition algorithm for event camera-based targets tracking, offering a reliable solution for various applications. [ABSTRACT FROM AUTHOR]

Existing neural network models directly add attention mechanisms to the network as a plug-and-play component to capture long-range dependencies and reconstruct feature maps. However, most methods do not fully tap the potential of attention in dealing with multi-scale problems. In this paper, an attention-guided YOLOv4 network (AG-YOLO) is proposed to address the multi-scale issue in object detection. We propose and apply multi-scale feature extraction to later stages of the backbone, which can not only enrich the feature hierarchy with low computational overhead, but also model the intra-scale and inter-scale correlation simultaneously to avoid missing key information. To reduce the redundant use of information flow, we propose a lightweight attention-guided feature pyramid network, which provides an efficient multi-level aggregation strategy based on multi-scale channel attention. In addition, a global context pathway is designed to reduce the dilution of high-level semantic information caused by information transmission. Compared with the baseline, AG-YOLO increased the mAP_0.5 by 1.67%, while the number of parameters and GFLOPs merely increased by 0.33M and 0.18, respectively. Meanwhile, the detection accuracy of small object categories has been improved. [ABSTRACT FROM AUTHOR]

Recent advances in multilayer perceptron (MLP) models have provided new effective network architecture designs for computer vision tasks. Compared with convolutional neural networks (CNNs) and visual transformers, MLP-based visual backbones have less induction bias, which can improve the sample utilization efficiency and reduce computational costs. Therefore, we designed the Mixer MLP Architecture for Image-Matching (MAIM), which is a coarse to fine-level detector-free image-matching scheme. Accordingly, we constructed a mixer MLP architecture called Mixer-WMLP, which evenly divides the feature map into non-overlapping windows, spreads each window as a token, achieves the exchange of token information between spatial locations, channels features through a two-layer MLP structure in the coarse-level model, and then feeds the windows with dense fine-level matching, thereby producing the final matches. Furthermore, the implemented global field-of-view mixer MLP framework for image-matching incurs a low computational cost. By conducting experiments with indoor and outdoor relative poses, our MLP architecture is compared with CNN and transformer-based image-matching methods. Our method has significant advantages in terms of real-time performance and largely reduces computational cost, proving its effectiveness in image-matching tasks. [ABSTRACT FROM AUTHOR]

Fault detection, isolation and recovery subsystems are accepted to make safety-critical systems resilient against faults and failures. Yet, these subsystems should be devised only for those faults that violate the system's requirements, while providing a correct approach such that requirements are met again. Consequently, the obtained system is minimal, although complete, and robust both with respect to safety and performance requirements. In this paper, we propose a systematic and automated approach based on formal methods that includes (1) the evaluation of the relevance of faults based on quantitative risk assessment, and (2) the validation of system robustness by statistical model checking. We apply this approach on an excerpt of a real-life autonomous robotics case study, and we report on the implementation and results obtained with the S BIP framework. [ABSTRACT FROM AUTHOR]

We present the first implementation of the monitoring of airborne fungal spores in real-time using digital holography. To obtain observations of Alternaria spp. spores representative of their airborne stage, we collected events measured in the air during crop harvesting in a contaminated potato field, using a Swisens Poleno device. The classification algorithm used by MeteoSwiss for operational pollen monitoring was extended by training the system using this additional dataset. The quality of the retrieved concentrations is evaluated by comparison with parallel measurements made with a manual Hirst-type trap. Correlations between the two measurements are high, especially over the main dispersion period of Alternaria spp., demonstrating the potential for automatic real-time monitoring of fungal spores. [ABSTRACT FROM AUTHOR]