Search

Image processing often requires filtering, which can be effectively performed by using B‐spline surface approximation. This article presents a fast method of approximating data in rectangular (image) form using such surfaces. The method considers dynamic changes of data by representing rows and columns in chordal parameterized form. To improve performance, after parametrization, lines are uniformly resampled using linear interpolation. It allows using the same basis functions with uniform parameterization for approximation of all processed lines. After approximation, reconstruction of original parameterization is required, but its computational complexity is also linear. [ABSTRACT FROM AUTHOR]

The purpose of long-term forecasting is to meet the needs of practical applications, such as the prediction of the development trend of infectious diseases and the planning of electricity consumption. In this paper, we study the long-term forecasting of time series. Studies have shown that previous Transformer-based models have the potential to improve prediction capabilities, but there are also some problems, such as the lack of location information and the slow training speed. To solve these problems, we designed an efficient Transformer-based model called AD-Autoformer, which is specifically designed for long-term series prediction. By introducing position embedding, the model can better understand the patterns and relationships in the sequence, to improve the performance and generalization ability of the model when processing sequence data, and the self-attention distilling mechanism realizes the compression and acceleration of the model by halving the cascading layer input to highlight the dominant attention. This method significantly reduces the computational complexity of the model and improves the training speed of the model while maintaining the performance of the model. Experimental results on five large datasets show that the proposed AD-Autoformer model has different degrees of improvement in MSE and MAE indicators compared with other benchmark methods. [ABSTRACT FROM AUTHOR]

The implementation of turn restrictions represents a critical research challenge in chiplet-based systems, with the objective of achieving deadlock-free communication. Nevertheless, existing methodologies encounter difficulties in terms of computational complexity, which impedes the design process. Moreover, as the scale of the problem increases, the cost of addressing it becomes increasingly untenable. In this paper, we introduce the efficient turn restrictions setting (ETRS) method to reduce the computational cost of implementing turn restrictions for boundary routers in chiplet-based systems. In stage 1, we present a symmetry-based preprocessing algorithm (SBPA). SBPA exploits the symmetry inherent in chiplet topologies by generating multiple sets of identical modes, ensuring that scenarios with the same objective function value are calculated only once per iteration. In stage 2, a heuristic selection algorithm (HSA) for turn restrictions based on NSGA-II is proposed as a means of searching for approximately optimal solutions and of solving the problem quickly. In stage 3, different filtering criteria are introduced to evaluate the Pareto fronts of HSA for making decisions on the placement of boundary routers and their turn restrictions. Evaluation results reveal that the proposed ETRS method surpasses existing solutions in terms of computation efficiency. Moreover, it delivers satisfactory optimal objective values. [ABSTRACT FROM AUTHOR]

Printed circuit boards (PCBs) play a critical role in electronic products. Ensuring these products' long-term reliability and consistent performance requires effective PCB defect detection. Although existing deep learning models for PCB defect detection are not highly accurate, they often neglect capability considerations. This paper introduces a precise, fast, and lightweight defect detection model, CCG-YOLO, based on an enhanced YOLOv5 model to address this issue. The enhancements in CCG-YOLO can be summarized as follows: (1) Improved Backbone network: The feature extraction ability of the Backbone network is enhanced by introducing a C3HB module, which fosters spatial interaction capabilities. (2) Lightweight feature fusion network: A lightweight convolution structure called Ghost-Shuffle Convolution is incorporated in the feature fusion network, remarkably reducing model parameters while maintaining performance. (3) Efficient residual networking: To enhance model performance further, a CNeB module is introduced based on the ConvNeXt network, which replaces the C3 module in the Neck. CNeB improves model detection accuracy and reduces the number of model parameters. The combination of these enhancements results in impressive performance. CCG-YOLO achieves mean average precision (mAP@0.5) of 99.5% and 88.75% in mAP@0.5:0.95 on the TDD-Net public dataset. Compared with the original YOLOv5s algorithm, CCG-YOLO offers a 4.24% improvement in mAP@0.5:0.95, a 1 MB reduction in model size, a 0.472 M decrease in the number of parameters, a 0.6G floating point operation reduction in computational complexity, and a 120 frames per second real-time inference speed. These experimental results underscore that the proposed model excels in accuracy and speed and has a compact size for PCB defect detection. Moreover, CCG-YOLO is easily deployable on low-end devices, making it well-suited for meeting the real-time requirements of industrial defect detection. [ABSTRACT FROM AUTHOR]

In this work, we propose a multiple feedback-based successive interference cancellation (SIC) scheme for an ultra-dense Internet of Things (IoT) device network. Non-orthogonal multiple access (NOMA) enables massive connectivity with improved user fairness and spectral efficiency and is envisaged as a multiple access technique for IoT devices. NOMA simultaneously serves multiple users within a single resource block, leading to unbounded yet regulated multi-user interference. SIC is widely adopted in the NOMA system to detect users' symbols. Nevertheless, multi-user interference and error propagation in the SIC layer are inherent challenges in NOMA. Recent studies have aimed to minimize interference and error propagation, imposing stringent conditions on the number of users and power allocation. Thus, this paper proposes novel multiple feedback-based SIC algorithms for the uplink multi-user NOMA scenarios that outperform the conventional SIC. Further, the proposed algorithm's performance is analyzed under the practical case of imperfect channel state information at the receiver node to validate the robustness. The computational complexity of multiple feedback SIC is compared with the conventional SIC. [ABSTRACT FROM AUTHOR]

In the development of web applications, the rapid advancement of Internet technologies has brought unprecedented opportunities and increased the demand for user authentication schemes. Before the emergence of blockchain technology, establishing trust between two unfamiliar entities relied on a trusted third party for identity verification. However, the failure or malicious behavior of such a trusted third party could undermine such authentication schemes (e.g., single points of failure, credential leaks). A secure authorization system is another requirement of user authentication schemes, as users must authorize other entities to act on their behalf in some situations. If the transfer of authentication permissions is not adequately restricted, security risks such as unauthorized transfer of permissions to entities may occur. Some research has proposed blockchain-based decentralized user authentication solutions to address these risks and enhance availability and auditability. However, as we know, most proposed schemes that allow users to transfer authentication permissions to other entities require significant gas consumption when deployed and triggered in smart contracts. To address this issue, we proposed an authentication scheme with transferability solely based on hash functions. By combining one-time passwords with Hashcash, the scheme can limit the number of times permissions can be transferred while ensuring security. Furthermore, due to its reliance solely on hash functions, our proposed authentication scheme has an absolute advantage regarding computational complexity and gas consumption in smart contracts. Additionally, we have deployed smart contracts on the Goerli test network and demonstrated the practicality and efficiency of this authentication scheme. [ABSTRACT FROM AUTHOR]

The unmanned aerial vehicle (UAV) city patrol is of great significance in ensuring the safety of residents' lives and properties, as well as maintaining the normal operation of the city. However, the detection of UAV images faces challenges such as numerous small-scale objects, complex backgrounds, and high requirements for detection speed. In response to these issues, we introduce a Real-time Small Object Detection network in UAV-vision (RTS-Net), tailored for UAV patrols. Initially, we introduce a multiscale feature fusion module (MFFM) designed to augment the expressiveness of features across scales, thereby enhancing the detection of smaller objects. Subsequently, leveraging attention mechanisms, we present the coordinated attention detection module (CADM), which bolsters the detection model's ability to accurately segregate objects from the background in expansive, complex scenarios. Lastly, a lightweight real-time feature extraction module (RFEM) is crafted to diminish model computational complexity and boost inference speed. On the UAV road patrol image dataset we constructed, our proposed method attains a detection accuracy of 89.9 % mAP, breaking previous records. It surpasses all prevailing detection methods, particularly for small-scale objects. Simultaneously, it achieves an inference speed of 163.9 FPS. The experimental results show that RTS-Net can satisfy the accurate and efficient detection of ground objects by various different UAV platforms in different complex scenarios. [ABSTRACT FROM AUTHOR]

Many compartments are prone to pose safety hazards such as loose fasteners or object intrusion due to their confined space, making manual inspection challenging. To address the challenges of complex inspection environments, diverse target categories, and variable scales in confined compartments, this paper proposes a novel GMS-YOLO network, based on the improved YOLOv8 framework. In addition to the lightweight design, this network accurately detects targets by leveraging more precise high-level and low-level feature representations obtained from GhostHGNetv2, which enhances feature-extraction capabilities. To handle the issue of complex environments, the backbone employs GhostHGNetv2 to capture more accurate high-level and low-level feature representations, facilitating better distinction between background and targets. In addition, this network significantly reduces both network parameter size and computational complexity. To address the issue of varying target scales, the first layer of the feature fusion module introduces Multi-Scale Convolutional Attention (MSCA) to capture multi-scale contextual information and guide the feature fusion process. A new lightweight detection head, Shared Convolutional Detection Head (SCDH), is designed to enable the model to achieve higher accuracy while being lighter. To evaluate the performance of this algorithm, a dataset for object detection in this scenario was constructed. The experiment results indicate that compared to the original model, the parameter number of the improved model decreased by 37.8%, the GFLOPs decreased by 27.7%, and the average accuracy increased from 82.7% to 85.0%. This validates the accuracy and applicability of the proposed GMS-YOLO network. [ABSTRACT FROM AUTHOR]

Simultaneous localization and mapping (SLAM) is an essential component for smart robot operations in unknown confined spaces such as indoors, tunnels and underground. This paper proposes a novel tightly-coupled ranging-LiDAR-inertial simultaneous localization and mapping framework, namely RLI-SLAM, which is designed to be high-accuracy, fast and robust in the long-term fast-motion scenario, and features two key innovations. The first one is tightly fusing the ultra-wideband (UWB) ranging and the inertial sensor to prevent the initial bias and long-term drift of the inertial sensor so that the point cloud distortion of the fast-moving LiDAR can be effectively compensated in real-time. This enables high-accuracy and robust state estimation in the long-term fast-motion scenario, even with a single ranging measurement. The second one is deploying an efficient loop closure detection module by using an incremental smoothing factor graph approach, which seamlessly integrates into the RLI-SLAM system, and enables high-precision mapping in a challenging environment. Extensive benchmark comparisons validate the superior accuracy of the proposed new state estimation and mapping framework over other state-of-the-art systems at a low computational complexity, even with a single ranging measurement and/or in a challenging environment. [ABSTRACT FROM AUTHOR]

Wi-Fi fingerprint-based indoor localization methods are effective in static environments but encounter challenges in dynamic, real-world scenarios due to evolving fingerprint patterns and feature spaces. This study investigates the temporal variations in signal strength over a 25-month period to enhance adaptive long-term Wi-Fi localization. Key aspects explored include the significance of signal features, the effects of sampling fluctuations, and overall accuracy measured by mean absolute error. Techniques such as mean-based feature selection, principal component analysis (PCA), and functional discriminant analysis (FDA) were employed to analyze signal features. The proposed algorithm, Ada-LT IP, which incorporates data reduction and transfer learning, shows improved accuracy compared to state-of-the-art methods evaluated in the study. Additionally, the study addresses multicollinearity through PCA and covariance analysis, revealing a reduction in computational complexity and enhanced accuracy for the proposed method, thereby providing valuable insights for improving adaptive long-term Wi-Fi indoor localization systems. [ABSTRACT FROM AUTHOR]

The use of linear array pushbroom images presents a new challenge in photogrammetric applications when it comes to transforming object coordinates to image coordinates. To address this issue, the Best Scanline Search/Determination (BSS/BSD) field focuses on obtaining the Exterior Orientation Parameters (EOPs) of each individual scanline. Current solutions are often impractical for real-time tasks due to their high time requirements and complexities. This is because they are based on the Collinearity Equation (CE) in an iterative procedure for each ground point. This study aims to develop a novel BSD framework that does not need repetitive usage of the CE with a lower computational complexity. The Linear Regression Model (LRM) forms the basis of the proposed BSD approach and uses Simulated Control Points (SCOPs) and Simulated Check Points (SCPs). The proposed method is comprised of two main steps: the training phase and the test phase. The SCOPs are used to calculate the unknown parameters of the LR model during the training phase. Then, the SCPs are used to evaluate the accuracy and execution time of the method through the test phase. The evaluation of the proposed method was conducted using ten various pushbroom images, 5 million SCPs, and a limited number of SCOPs. The Root Mean Square Error (RMSE) was found to be in the order of ten to the power of negative nine (pixel), indicating very high accuracy. Furthermore, the proposed approach is more robust than the previous well-known BSS/BSD methods when handling various pushbroom images, making it suitable for practical and real-time applications due to its high speed, which only requires 2–3 s of time. [ABSTRACT FROM AUTHOR]

Accurate estimation of State-of-Charge (SoC) is essential for ensuring the safe and efficient operation of electric vehicles (EVs). Currently, second-order RC equivalent circuit models do not account for the influence of battery charging and discharging states on battery parameters. Additionally, offline parameter identification becomes inaccurate as the battery ages. Online identification requires real-time parameter updates during the SoC estimation process, which increases the computational complexity and reduces the computational efficiency of real vehicle Battery Management System (BMS) chips. To address these issues, this paper proposes a SoC estimation method that combines online and offline identification based on an optimized second-order RC equivalent circuit model, which distinguishes it from existing methods in the field. On the basis of the traditional second-order RC model, the Ohmic resistance (R0), polarization resistance (R1), polarization capacitance (C1), diffusion resistance (R2), and diffusion capacitance (C2) during the charging and discharging processes are discussed separately. R0, which does not change frequently, is identified offline, while R1, R2, C1, and C2, which dynamically change with time and current, are identified online. To thoroughly verify the feasibility of the proposed method, we construct an SoC estimation test bench, which allows us to adjust the battery's surface temperature in real time using a temperature control chamber. Experimental validation under Federal Urban Driving Schedule (FUDS) (−10 °C to 45 °C, 80% battery capacity) and Dynamic Stress Test (DST) (−10 °C to 45 °C, 8% battery capacity) conditions demonstrate that our method improves SoC estimation accuracy by 16.28% under FUDS and 28.2% under DST compared to the improved GRU-based transfer learning method, while maintaining system SoC estimation efficiency. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) inter-pulse parameter agility technology involves dynamically adjusting parameters such as the pulse width, chirp rate, carrier frequency, and pulse repetition interval within a certain range; this effectively increases the complexity and uncertainty of radar waveforms, thereby countering active deceptive interference signals from multiple dimensions. With the development of active deceptive interference technology, single-parameter agility can no longer meet the requirements, making multi-parameter joint agility one of the main research directions. However, inter-pulse carrier frequency agility can cause azimuth Doppler chirp rate variation, making azimuth compression difficult and compensation computationally intensive, thus hindering imaging. Additionally, pulse repetition interval (PRI) agility leads to non-uniform azimuth sampling, severely deteriorating image quality. To address these issues, this paper proposes a multi-parameter agile SAR imaging scheme based on traditional frequency domain imaging algorithms. This scheme can handle joint agility of pulse width, chirp rate polarity, carrier frequency, and PRI, with relatively low computational complexity, making it feasible for engineering implementation. By inverting SAR images, the echoes with multi-parameter joint agility are obtained, and active deceptive interference signals are added for processing. The interference-suppressed imaging results verify the effectiveness of the proposed method. Furthermore, simulation results of point targets with multiple parameters under the proposed processing algorithm show that the peak sidelobe ratio (PSLR) and integrated sidelobe ratio (ISLR) are improved by 12 dB and 10 dB, respectively, compared to the traditional fixed waveform scheme. [ABSTRACT FROM AUTHOR]

The utilization of Synthetic Aperture Radar (SAR) for real-time ship detection proves highly advantageous in the supervision and monitoring of maritime activities. Ship detection in the range-compressed domain of SAR rather than in fully focused SAR imagery can significantly reduce the time and computational resources required for complete SAR imaging, enabling lightweight real-time ship detection methods to be implemented on an airborne or spaceborne SAR platform. However, there is a lack of lightweight ship detection methods specifically designed for the SAR range-compressed domain. In this paper, we propose Fast Range-Compressed Detection (FastRCDet), a novel lightweight network for ship detection in the SAR range-compressed domain. Firstly, to address the distinctive geometric characteristics of the SAR range-compressed domain, we propose a Lightweight Adaptive Network (LANet) as the backbone of the network. We introduce Arbitrary Kernel Convolution (AKConv) as a fundamental component, which enables the flexible adjustment of the receptive field shape and better adaptation to the large scale and aspect ratio characteristics of ships in the range-compressed domain. Secondly, to enhance the efficiency and simplicity of the network model further, we propose an innovative Multi-Scale Fusion Head (MSFH) module directly integrated after the backbone, eliminating the need for a neck module. This module effectively integrates features at various scales to more accurately capture detailed information about the target. Thirdly, to further enhance the network's adaptability to ships in the range-compressed domain, we propose a novel Direction IoU (DIoU) loss function that leverages angle cost to control the convergence direction of predicted bounding boxes, thereby improving detection accuracy. Experimental results on a publicly available dataset demonstrate that FastRCDet achieves significant reductions in parameters and computational complexity compared to mainstream networks without compromising detection performance in SAR range-compressed images. FastRCDet achieves a low parameter of 2.49 M and a high detection speed of 38.02 frames per second (FPS), surpassing existing lightweight detection methods in terms of both model size and processing rate. Simultaneously, it attains an average accuracy (AP) of 77.12% in terms of its detection performance. This method provides a baseline in lightweight network design for SAR ship detection in the range-compressed domain and offers practical implications for resource-constrained embedded platforms. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have significantly advanced in recent years in detecting arbitrary-oriented ships in synthetic aperture radar (SAR) images. However, challenges remain with multi-scale target detection and deployment on satellite-based platforms due to the extensive model parameters and high computational complexity. To address these issues, we propose a lightweight method for arbitrary-oriented ship detection in SAR images, named LSR-Det. Specifically, we introduce a lightweight backbone network based on contour guidance, which reduces the number of parameters while maintaining excellent feature extraction capability. Additionally, a lightweight adaptive feature pyramid network is designed to enhance the fusion capability of the ship features across different layers with a low computational cost by incorporating adaptive ship feature fusion modules between the feature layers. To efficiently utilize the fused features, a lightweight rotating detection head is designed, incorporating the idea of sharing the convolutional parameters, thereby improving the network's ability to detect multi-scale ship targets. The experiments conducted on the SAR ship detection dataset (SSDD) and the rotating ship detection dataset (RSDD-SAR) demonstrate that LSR-Det achieves an average precision (AP50) of 98.5% and 97.2% with 3.21 G floating point operations (FLOPs) and 0.98 M parameters, respectively, outperforming the current popular SAR arbitrary-direction ship target detection methods. [ABSTRACT FROM AUTHOR]

Hand pose estimation is a fundamental task in computer vision with applications in virtual reality, gesture recognition, autonomous driving, and virtual surgery. Keypoint detection often relies on deep learning methods and high-resolution feature map representations to achieve accurate detection. The HRNet framework serves as the basis, but it presents challenges in terms of extensive parameter count and demanding computational complexity due to high-resolution representations. To mitigate these challenges, we propose a lightweight keypoint detection network called LGCANet (Lightweight Ghost-Coordinate Attention Network). This network primarily consists of a lightweight feature extraction head for initial feature extraction and multiple lightweight foundational network modules called GCAblocks. GCAblocks introduce linear transformations to generate redundant feature maps while concurrently considering inter-channel relationships and long-range positional information using a coordinate attention mechanism. Validation on the RHD dataset and the COCO-WholeBody-Hand dataset shows that LGCANet reduces the number of parameters by 65.9% and GFLOPs by 72.6% while preserving the accuracy and improves the detection speed. [ABSTRACT FROM AUTHOR]

Lunar missions are currently experiencing a significant surge in popularity, presenting expansive opportunities for further exploration and development. To thoroughly explore the design margins and potential of lunar landers, and to foster the development of overall designs driven by comprehensive performance objectives, it is crucial to conduct optimization design considering the coupling between key disciplines such as trajectory and propulsion. Considering the significant increase in computational complexity caused by conducting trajectory/propulsion integrated design optimization, the analytical target cascading method is employed to hierarchically decompose and coordinate optimization of the complex systems. This article presents a phased soft-landing strategy on the manned lunar lander propelled by hybrid rocket motors, utilizing powered explicit guidance and Apollo powered descent guidance, and proceeds with the trajectory/propulsion integrated design optimization involving diverse grain shapes and feed systems. This optimization process is separately undertaken utilizing multidisciplinary feasible method and analytical target cascading method. The analysis reveals that integrating trajectory and propulsion considerations into the optimization process facilitates a 5 % reduction in the overall mass relative to optimizations constrained solely by velocity increment and lack comprehensive trajectory design considerations. This highlights the profound impact of trajectory requirements on propulsion system design and the advantages of powered explicit guidance laws in minimizing fuel consumption. Crucially, the use of analytical target cascading achieves the better optimization results, and significantly reduces subsystem evaluation times, enhancing operational efficiency by 48 %, demonstrating the advantage in handling complex, large-scale systems. On another level, with different β values, the Mean Relative Error of the target values for the three schemes obtained by the analytical target cascading method is 0.0016, indicating good stability and strong robustness. The practical exploration in this article provides methods and frameworks for high-performance optimization design of complex aerospace mission profiles in the future. • The trajectory/propulsion integrated design optimization is conducted. • The two-phase guidance strategy for lunar lander descent is proposed. • Based on the strategy, the analytical target cascading-decomposed framework is set. • The influence of trajectory and propulsion coupling on overall design is revealed. • The optimization effectiveness, convergence rate, and robustness of the ATC is shown. [ABSTRACT FROM AUTHOR]

Model order reduction technique provides an effective way to reduce computational complexity in large-scale circuit simulations. This paper proposes a new model order reduction method for delay circuit systems based on Hermite expansion technique. The presented method consists of three steps i.e., first the delay elements are approximated using the recursive relation of Hermite polynomials, then in the second step, the reduced order is estimated for the delay circuit system using a delay truncation in the Hermite domain and in the third step, a multi-order Arnoldi process is computed for obtaining the projection matrix. In the following, the reduced order delay circuit model is obtained by the projection matrix. Moment matching and passivity properties of the reduced circuit system are also analyzed. Two circuit examples with delay components are performed to verify the effectiveness of the proposed MOR approach. [ABSTRACT FROM AUTHOR]

Electronic equipment used on a daily basis now frequently includes active noise cancellation. The adaptive filters, which are positioned within, are essential for noise cancellation. An essential component to take into account for the overall performance is the structural and computational complexity of the filter. The filter's structure has an impact on this. The amount of taps determines the structure. Active noise cancellation filters often have set tap lengths and are lengthy, which causes sluggish convergence and delay. As a result, a trade-off between the filter's length and convergence is required. This is conceivable if there is a flexible filter with a tap length that adapts to the environment while still ensuring acceptable convergence. This study proposes a novel Minimum Mean M-estimate method with changeable tap length and uses a sub-band adaptive filtering technique to shorten the filter's length. In order to maximize the filter's efficiency, the advantages of three approaches are specifically merged in this work. They are the proposed algorithm, the proposed method's variable tap length variant, and the sub-band adaptive filtering. The simulation's findings and recommendations are supported. [ABSTRACT FROM AUTHOR]

Speech enhancement performance has improved significantly with the introduction of deep learning models, especially methods based on the Long–Short-Term Memory architecture. However, these methods face challenges such as high computational complexity and redundancy of input features. To address these issues, we propose a U-Net-based approach that utilizes an encoder/decoder to extract more concise features, thereby enhancing single-channel speech performance and reducing computation complexity. The proposed method includes a Cross-Swin-Transformer block and a masked bottleneck module, which down-samples features while preserving the detailed representation through skip connections and carefully designed blocks. The bottleneck module extracts coarse representations of hidden features as masks. We evaluated our method against other U-Net-based approaches on VCTK and DNS corpora using CBAK, eSTOI, PESQ, STOI, and SI-SDR metrics. The results demonstrate that the proposed method achieves promising performance while significantly reducing computational complexity. [ABSTRACT FROM AUTHOR]

The multi-state flow network (MSFN) serves as a fundamental framework for real-life network-structured systems and various applications. The system reliability of the MSFN, denoted as Rd, is defined as the probability of successfully transmitting at least d units of demand from a source to a terminal. Current analytical algorithms are characterized by their computational complexity, specifically falling into the NP-hard problem to evaluate exact system reliability. Moreover, existing analytical algorithms for calculating Rd are basically designed for predetermined values of d. This limitation hinders the ability of decision-makers to flexibly choose the most appropriate based on the specific characteristics of the given scenarios or applications. This means that these methods are incapable of simultaneously calculating system reliability for various demand levels. Therefore, this paper develops a simulation-based algorithm to estimate system reliability for all possible demand levels simultaneously such that we can eliminate the need to rely on repeat procedures for each specified d. An experimental investigation was carried out on a benchmark network and a practical network to validate the effectiveness and performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Graph matching is a commonly used technique in computer vision and pattern recognition. Recent data-driven approaches have improved the graph matching accuracy remarkably, whereas some traditional algorithm-based methods are more robust to feature noises, outlier nodes, and global transformation (e.g. rotation). In this paper, we propose a graph neural network (GNN) based approach to combine the advantage of data-driven and traditional methods. In the GNN framework, we transform traditional graph matching solvers as single-channel GNNs on the association graph and extend the single-channel architecture to the multi-channel network. The proposed model can be seen as an ensemble method that fuses multiple algorithms at every iteration. Instead of averaging the estimates at the end of the ensemble, in our approach, the independent iterations of the ensembled algorithms exchange their information after each iteration via a 1 × 1 channel-wise convolution layer. Experiments show that our model improves the performance of traditional algorithms significantly. In addition, we propose a random sampling strategy to reduce the computational complexity and GPU memory usage, so that the model is applicable to matching graphs with thousands of nodes. We evaluate the performance of our method on three tasks: geometric graph matching, semantic feature matching, and few-shot 3D shape classification. The proposed model performs comparably or outperforms the best existing GNN-based methods. [ABSTRACT FROM AUTHOR]

Existing sea surface small target detection methods typically rely on intricate feature extraction techniques on transformed radar returns. However, these approaches suffer from issues of high computational complexity and low real-time performance. Temporal Convolutional Network (TCN) can enable direct processing of radar time-series echo data without the need for elaborate feature extraction, thus substantially improving computational efficiency. Building upon this, this paper presents a novel target detection algorithm based on Multi-layer Attention Temporal Convolutional Network (MA-TCN). The proposed algorithm processes the amplitude information in the original echo signals, and comprehensively extracts sequence feature information through the construction of stacked residual modules. Additionally, it integrates multi-layer attention mechanisms to adaptively adjust the output weights of each residual module, thereby further enhancing detection accuracy. Experimental results demonstrate that the proposed approach achieves significant improvements in both detection performance and efficiency compared to existing methods. [ABSTRACT FROM AUTHOR]

In recent years, the utilization of unmanned aerial vehicles (UAVs) for aerial target detection has gained significant attention due to their high-altitude perspective and maneuverability, which offer novel opportunities and tremendous potential in this field. However, detecting targets in UAV aerial images remains highly challenging due to the presence of numerous small targets with limited feature information, as well as issues like target occlusion and complex backgrounds that severely impact detection accuracy. To address these challenges, we propose a detection model called BDC-YOLOv8 that aims to enhance accuracy for small targets while minimizing computational complexity. Specifically, we augment the YOLOv8 architecture by incorporating a dedicated detection head tailored for small targets to improve performance when encountering such objects. Additionally, we restructure the neck network of the model to better extract and fuse feature information from targets with significant scale variations. Furthermore, we introduce the concept of DynamicHead to enhance the detection head by incorporating various attention mechanisms suitable for our task ahead of the original detection head, thereby enhancing the model's capability to detect objects of different scales and complex backgrounds. Moreover, we introduce Convolutional Block Attention Module (CBAM) to identify regions of interest in densely populated areas. Extensive experiments conducted on the VisDrone2019 dataset yield promising results where our model achieves a mean Average Precision (mAP) score of 38% and an AP50 score of 59.6%. Compared to the original YOLOV8 model, improvements are observed with increases in mAP by 2.5% and AP50 by 3.7%, respectively. Notably, our model demonstrates a significant enhancement in detecting small targets with an increase in APs evaluation metric by 4.1%. [ABSTRACT FROM AUTHOR]

This work investigates the less conservative stability conditions for linear systems with a time-varying delay. At first, augmented Lyapunov–Krasovskii functionals(LKFs) are constructed with state vectors that have not been utilized in the existing works, and an augmented zero equality that can be derived according to the augmented vector is proposed. By utilizing them, a stability condition is proposed in the form of a linear matrix inequality. And, by using novel delay-dependent LKFs and the introduced ones, improved results are obtained than the previous result. The addition of the delay-dependent LKFs increases the number of decision variables in the results. Therefore, any vectors of integral inequalities utilized in the proposed criterion are appropriately adjusted to reduce computational complexity. To check the excellence and validity of the proposed results, several numerical examples are applied. • The improved stability conditions for the linear system with a time-varying delay are proposed. • The proposed Lyapunov-Krasovskii functionals(LKFs) have more extended augmented vectors to obtain less conservative results. And LKFs based on an integral inequality (Lee et al., 2023) is utilized. • The novel LKFs which are the delay-dependent forms are proposed. The superiority of the results is checked according to the presence or absence of the forms. • The computational complexity is reduced by adjusting any vector of the integral inequality utilized in the proposed criteria. [ABSTRACT FROM AUTHOR]

The parallel mechanism exhibits high stiffness and excellent dynamic response, making it ideal for high-precision applications. In our early work, a novel 6-DOF redundant parallel posture mechanism with four limbs for solar wing docking has been proposed; each limb consists of three links and four joints. This paper primarily focuses on optimization design of the mechanism. The calculation of workspace volume reveals that factors influencing the range of posture adjustment include dynamic platform parameters, static platform parameters, the drive trajectory of each kinematic pair, and the angles between each kinematic pair. A sensitivity analysis was conducted to examine the impact of each parameter on the range of posture adjustment. To reduce computational complexity and improve analysis efficiency, a combined approach of single-factor analysis and response surface methodology (RSM) is used in the paper. Single-factor analysis is utilized to evaluate the effect of each parameter on the posture adjustment range. Based on these results, RSM is used to establish a regression model for parameters; thereby, the optimal parameter combination for the mechanism is determined. The regression coefficient R2 = 0.9374 attests to the validity of the proposed model. Finally, a comparison of the posture adjustment range before and after optimization is presented, providing a foundation for the practical application of the redundant parallel mechanism. This paper introduces a novel structural design concept aimed at resolving the conflict between heavy loads and compact sizes in redundant parallel mechanisms while providing valuable insights for miniaturized design. [ABSTRACT FROM AUTHOR]

Sea fog reduces visibility to less than 1 km and is a major cause of maritime accidents, particularly affecting the navigation of small fishing vessels as it forms when warm, moist air moves over cold water, making it difficult to predict. Traditional visibility measurement tools are costly and limited in their real-time monitoring capabilities, which has led to the development of video-based algorithms using cameras. This study introduces the Approximating and Eliminating the Airlight–Reduced DCP (AEA-RDCP) algorithm, designed to address the issue where sunlight reflections are mistakenly recognized as fog in existing video-based sea fog intensity measurement algorithms, thereby improving performance. The dataset used in the experiment is categorized into two types: one consisting of images unaffected by sunlight and another consisting of maritime images heavily influenced by sunlight. The AEA-RDCP algorithm enhances the previously researched RDCP algorithm by effectively eliminating the influence of atmospheric light, utilizing the initial stages of the Dark Channel Prior (DCP) process to generate the Dark Channel image. While the DCP algorithm is typically used for dehazing, this study employs it only to the point of generating the Dark Channel, reducing computational complexity. The generated image is then used to estimate visibility based on a threshold for fog density estimation, maintaining accuracy while reducing computational demands, thereby allowing for the real-time monitoring of sea conditions, enhancing maritime safety, and preventing accidents. [ABSTRACT FROM AUTHOR]

Wearing safety helmets at construction sites is a major measure to prevent safety accidents, so it is essential to supervise and ensure that workers wear safety helmets. This requires a high degree of real-time performance. We improved the network structure based on YOLOv7. To enhance real-time performance, we introduced GhostModule after comparing various modules to create a new efficient structure that generates more feature mappings with fewer linear operations. SE blocks were introduced after comparing several attention mechanisms to highlight important information in the image. The EIOU loss function was introduced to speed up the convergence of the model. Eventually, we constructed the efficient model EGS-YOLO. EGS-YOLO achieves a mAP of 91.1%, 0.2% higher than YOLOv7, and the inference time is 13.3% faster than YOLOv7 at 3.9 ms (RTX 3090). The parameters and computational complexity are reduced by 37.3% and 33.8%, respectively. The enhanced real-time performance while maintaining the original high precision can meet actual detection requirements. [ABSTRACT FROM AUTHOR]

In response to the challenges of small-size defects and low recognition rates in Printed Circuit Boards (PCBs), as well as the need for lightweight detection models that can be embedded in portable devices, this paper proposes an improved defect detection method based on a lightweight shared convolutional head using YOLOv8n. Firstly, the Spatial and Channel reconstruction Convolution (SCConv) is embedded into the Cross Stage Partial with Convolutional Layer Fusion (C2f) structure of the backbone network, which reduces redundant computations and enhances the model's learning capacity. Secondly, an adaptive feature selection module is integrated to improve the network's ability to recognize small targets. Subsequently, a Shared Lightweight Convolutional Detection (SLCD) Head replaces the original Decoupled Head, reducing the model's computational complexity while increasing detection accuracy. Finally, the Weighted Intersection over Union (WIoU) loss function is introduced to provide more precise evaluation results and improve generalization capability. Comparative experiments conducted on a public PCB dataset demonstrate that the improved algorithm achieves a mean Average Precision (mAP) of 98.6% and an accuracy of 99.8%, representing improvements of 3.8% and 3.1%, respectively, over the original model. The model size is 4.1 M, and its FPS is 144.1, meeting the requirements for real-time and lightweight portable deployment. [ABSTRACT FROM AUTHOR]

Printed circuit boards present several challenges to the detection of defects, including targets of insufficient size and distribution, a high level of background noise, and a variety of complex types. These factors contribute to the difficulties encountered by PCB defect detection networks in accurately identifying defects. This paper proposes a less-parametric model, YOLO-RRL, based on the improved YOLOv8 architecture. The YOLO-RRL model incorporates four key improvement modules: The following modules have been incorporated into the proposed model: Robust Feature Downsampling (RFD), Reparameterised Generalised FPN (RepGFPN), Dynamic Upsampler (DySample), and Lightweight Asymmetric Detection Head (LADH-Head). The results of multiple performance metrics evaluation demonstrate that YOLO-RRL enhances the mean accuracy (mAP) by 2.2 percentage points to 95.2%, increases the frame rate (FPS) by 12%, and significantly reduces the number of parameters and the computational complexity, thereby achieving a balance between performance and efficiency. Two datasets, NEU-DET and APSPC, were employed to evaluate the performance of YOLO-RRL. The results indicate that YOLO-RRL exhibits good adaptability. In comparison to existing mainstream inspection models, YOLO-RRL is also more advanced. The YOLO-RRL model is capable of significantly improving production quality and reducing production costs in practical applications while also extending the scope of the inspection system to a wide range of industrial applications. [ABSTRACT FROM AUTHOR]

Free speech is essential, but it can conflict with protecting marginalized groups from harm caused by hate speech. Social media platforms have become breeding grounds for this harmful content. While studies exist to detect hate speech, there are significant research gaps. First, most studies used text data instead of other modalities such as videos or audio. Second, most studies explored traditional machine learning algorithms. However, due to the increase in complexities of computational tasks, there is need to employ complex techniques and methodologies. Third, majority of the research studies have either been evaluated using very few evaluation metrics or not statistically evaluated at all. Lastly, due to the opaque, black-box nature of the complex classifiers, there is need to use explainability techniques. This research aims to address these gaps by detecting hate speech in English and Kiswahili languages using videos manually collected from YouTube. The videos were converted to text and used to train various classifiers. The performance of these classifiers was evaluated using various evaluation and statistical measurements. The experimental results suggest that the random forest classifier achieved the highest results for both languages across all evaluation measurements compared to all classifiers used. The results for English language were: accuracy 98%, AUC 96%, precision 99%, recall 97%, F1 98%, specificity 98% and MCC 96% while the results for Kiswahili language were: accuracy 90%, AUC 94%, precision 93%, recall 92%, F1 94%, specificity 87% and MCC 75%. These results suggest that the random forest classifier is robust, effective and efficient in detecting hate speech in any language. This also implies that the classifier is reliable in detecting hate speech and other related problems in social media. However, to understand the classifiers' decision-making process, we used the Local Interpretable Model-agnostic Explanations (LIME) technique to explain the predictions achieved by the random forest classifier. [ABSTRACT FROM AUTHOR]

Several studies have attempted to solve traveling salesman problems (TSPs) using various deep learning techniques. Among them, Transformer-based models show state-of-the-art performance even for large-scale Traveling Salesman Problems (TSPs). However, they are based on fully-connected attention models and suffer from large computational complexity and GPU memory usage. Our work is the first CNN-Transformer model based on a CNN embedding layer and partial self-attention for TSP. Our CNN-Transformer model is able to better learn spatial features from input data using a CNN embedding layer compared with the standard Transformer-based models. It also removes considerable redundancy in fully-connected attention models using the proposed partial self-attention. Experimental results show that the proposed CNN embedding layer and partial self-attention are very effective in improving performance and computational complexity. The proposed model exhibits the best performance in real-world datasets and outperforms other existing state-of-the-art (SOTA) Transformer-based models in various aspects. Our code is publicly available at https://github.com/cm8908/CNN_Transformer3. [ABSTRACT FROM AUTHOR]

Facial emotion recognition has become a complicated task due to individual variations in facial characteristics, as well as racial and cultural variances. Different psychological studies show that there are complex expressions other than basic emotions which are made up of two basic emotions like"Happily Disgusted", "Happily Surprised", "Sadly Surprised", etc. Compound emotion recognition is challenging due to very less publicly available compound emotion datasets which are imbalanced too. In this paper, we have proposed an LSwin-CBAM for the classification of compound emotions. To address the problem of the imbalanced dataset, the proposed model exploits the cutmix augmentation technique for data augmentation. It also incorporates the CBAM attention mechanism to emphasize the relevant features in an image and swin transformer with fewer swin transformer blocks which leads to less computational complexity in terms of trainable parameters and improves the overall classification accuracy as well. The experimental results of LSwin-CBAM on RAF-DB and EmotioNet datasets show that the proposed transformer-based network can well recognize compound emotions. [ABSTRACT FROM AUTHOR]

To accelerate the solution of electromagnetic scattering problems, compressive sensing (CS) has been introduced into the method of moments (MoM). Consequently, a computational model based on underdetermined equations has been proposed, which effectively reduces the computational complexity compared with the traditional MoM. However, while solving surface-integral formulations for three-dimensional targets by MoM, due to the severe oscillation of current signals, commonly used sparse bases become inapplicable, which renders the application of the underdetermined equation model quite challenging. To address this issue, this paper puts forward a scheme that employs Krylov subspace, which is constructed with low complexity by meticulously designing a group of non-orthogonal basis vectors, to replace the sparse transforms in the algorithmic framework. The principle of the method is elaborated in detail, and its effectiveness is validated through numerical experiments. [ABSTRACT FROM AUTHOR]

Addressing the challenges in effectively extracting multi-scale features and preserving pose information during hyperspectral image (HSI) classification, a Multi-Scale Depthwise Separable Capsule Network (MDSC-Net) is proposed in this article for HSI classification. Initially, hierarchical features are extracted by MDSC-Net through the employment of parallel multi-scale convolutional kernels, while computational complexity is reduced via depthwise separable convolutions, thus reducing the overall computational load and achieving efficient feature extraction. Subsequently, to enhance the translational invariance of features and reduce the loss of pose information, features of various scales are processed in parallel by independent capsule networks, with improvements in max pooling achieved through dynamic routing. Lastly, features of different scales are concatenated and integrated through the concatenate operation, thereby facilitating precise analysis of multi-level information in the hyperspectral image classification process. Experimental comparisons demonstrate that MDSC-Net achieves average accuracies of 94%, 98%, and 99% on the Kennedy Space Center, University of Pavia, and Salinas datasets, respectively, indicating a significant performance advantage over recent HSI classification models and validating the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

Copyright of Zhejiang Electric Power is the property of Zhejiang Electric Power Editorial Office 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.)

As one of the most important economic crops, grapes have attracted considerable attention due to their high yield, rich nutritional value, and various health benefits. Identifying grape bunches is crucial for maintaining the quality and quantity of grapes, as well as managing pests and diseases. In recent years, the combination of automated equipment with object detection technology has been instrumental in achieving this. However, existing lightweight object detection algorithms often sacrifice detection precision for processing speed, which may pose obstacles in practical applications. Therefore, this thesis proposes a lightweight detection method named YOLOv8s-grape, which incorporates several effective improvement points, including modified efficient channel attention (MECA), slim-neck, new spatial pyramid pooling fast (NSPPF), dynamic upsampler (DySample), and intersection over union with minimum point distance (MPDIoU). In the proposed method, MECA and NSPPF enhance the feature extraction capability of the backbone, enabling it to better capture crucial information. Slim-neck reduces redundant features, lowers computational complexity, and effectively reuses shallow features to obtain more detailed information, further improving detection precision. DySample achieves excellent performance while maintaining lower computational costs, thus demonstrating high practicality and rapid detection capability. MPDIoU enhances detection precision through faster convergence and more precise regression results. Experimental results show that compared to other methods, this approach performs better in the grapevine bunch detection dataset and grapevine bunch condition detection dataset, with mean average precision (mAP50-95) increasing by 2.4% and 2.6% compared to YOLOv8s, respectively. Meanwhile, the computational complexity and parameters of the method are also reduced, with a decrease of 2.3 Giga floating-point operations per second and 1.5 million parameters. Therefore, it can be concluded that the proposed method, which integrates these improvements, achieves lightweight and highprecision detection, demonstrating its effectiveness in identifying grape bunches and assessing biophysical anomalies. [ABSTRACT FROM AUTHOR]

Load prediction tasks aim to predict the dynamic trend of future load based on historical performance sequences, which are crucial for cloud platforms to make timely and reasonable task scheduling. However, existing prediction models are limited while capturing complicated temporal patterns from the load sequences. Besides, the frequently adopted global weighting strategy (e.g., the self-attention mechanism) in temporal modeling schemes has quadratic computational complexity, hindering the immediate response of cloud servers in complex real-time scenarios. To address the above limitations, we propose a Wavelet decomposition-enhanced External Transformer (WETformer) to provide accurate yet efficient load prediction for cloud servers. Specifically, we first incorporate discrete wavelet transform to progressively extract long-term trends, highlighting the intrinsic attributes of temporal sequences. Then, we propose a lightweight multi-head External Attention (EA) mechanism to simultaneously consider the inter-element relationships within load sequences and the correlations across different sequences. Such an external component has linear computational complexity, mitigating the encoding redundancy prevalent and enhancing prediction efficiency. Extensive experiments conducted on Alibaba Cloud's cluster tracking dataset demonstrate that WETformer achieves superior prediction accuracy and the shortest inference time compared to several state-of-the-art baseline methods. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) is a crucial task in various applications, including healthcare, fitness, and the military. Deep learning models have revolutionized HAR, however, their computational complexity, particularly those involving BiLSTMs, poses significant challenges for deployment on resource-constrained devices like smartphones. While BiLSTMs effectively capture long-term dependencies by processing inputs bidirectionally, their high parameter count and computational demands hinder practical applications in real-time HAR. This study investigates the approximation of the computationally intensive BiLSTM component in a HAR model by using a combination of alternative model components and data flipping augmentation. The proposed modifications to an existing hybrid model architecture replace the BiLSTM with standard and residual LSTM, along with convolutional networks, supplemented by data flipping augmentation to replicate the context awareness typically provided by BiLSTM networks. The results demonstrate that the residual LSTM (ResLSTM) model achieves superior performance while maintaining a lower computational complexity compared to the traditional BiLSTM model. Specifically, on the UCI-HAR dataset, the ResLSTM model attains an accuracy of 96.34% with 576,702 parameters, outperforming the BiLSTM model's accuracy of 95.22% with 849,534 parameters. On the WISDM dataset, the ResLSTM achieves an accuracy of 97.20% with 192,238 parameters, compared to the BiLSTM's 97.23% accuracy with 283,182 parameters, demonstrating a more efficient architecture with minimal performance trade-off. For the KU-HAR dataset, the ResLSTM model achieves an accuracy of 97.05% with 386,038 parameters, showing comparable performance to the BiLSTM model's 98.63% accuracy with 569,462 parameters, but with significantly fewer parameters. [ABSTRACT FROM AUTHOR]

In infrared detection scenarios, detecting and recognizing low-contrast and small-sized targets has always been a challenge in the field of computer vision, particularly in complex road traffic environments. Traditional target detection methods usually perform poorly when processing infrared small targets, mainly due to their inability to effectively extract key features and the significant feature loss that occurs during feature transmission. To address these issues, this paper proposes a fast detection and recognition model based on a multi-scale self-attention mechanism, specifically for small road targets in infrared detection scenarios. We first introduce and improve the DyHead structure based on the YOLOv8 algorithm, which employs a multi-head self-attention mechanism to capture target features at various scales and enhance the model's perception of small targets. Additionally, to prevent information loss during the feature transmission process via the FPN structure in traditional YOLO algorithms, this paper introduces and enhances the Gather-and-Distribute Mechanism. By computing dependencies between features using self-attention, it reallocates attention weights in the feature maps to highlight important features and suppress irrelevant information. These improvements significantly enhance the model's capability to detect small targets. Moreover, to further increase detection speed, we pruned the network architecture to reduce computational complexity and parameter count, making the model suitable for real-time processing scenarios. Experiments on our self built infrared road traffic dataset (mainly including two types of targets: vehicles and people) show that compared with the baseline, our method achieves a 3.1% improvement in AP and a 2.5% increase in mAP on the VisDrone2019 dataset, showing significant enhancements in both detection accuracy and processing speed for small targets, with improved robustness and adaptability. [ABSTRACT FROM AUTHOR]

Radar signal intra-pulse modulation recognition can be addressed with convolutional neural networks (CNNs) and time–frequency images (TFIs). However, current CNNs have high computational complexity and do not perform well in low-signal-to-noise ratio (SNR) scenarios. In this paper, we propose a lightweight CNN known as the cross-scale aware network (CSANet) to recognize intra-pulse modulation based on three types of TFIs. The cross-scale aware (CSA) module, designed as a residual and parallel architecture, comprises a depthwise dilated convolution group (DDConv Group), a cross-channel interaction (CCI) mechanism, and spatial information focus (SIF). DDConv Group produces multiple-scale features with a dynamic receptive field, CCI fuses the features and mitigates noise in multiple channels, and SIF is aware of the cross-scale details of TFI structures. Furthermore, we develop a novel time–frequency fusion (TFF) feature based on three types of TFIs by employing image preprocessing techniques, i.e., adaptive binarization, morphological processing, and feature fusion. Experiments demonstrate that CSANet achieves higher accuracy with our TFF compared to other TFIs. Meanwhile, CSANet outperforms cutting-edge networks across twelve radar signal datasets, providing an efficient solution for high-precision recognition in low-SNR scenarios. [ABSTRACT FROM AUTHOR]

Multiple dynamic impact signals are widely used in a variety of engineering scenarios and are difficult to identify accurately and quickly due to the signal adhesion phenomenon caused by nonlinear interference. To address this problem, an intelligent algorithm combining wavelet transforms with lightweight neural networks is proposed. First, the features of multiple impact signals are analyzed by establishing a transfer model for multiple impacts in multibody dynamical systems, and interference is suppressed using wavelet transformation. Second, a lightweight neural network, i.e., fast-activated minimal gated unit (FMGU), is elaborated for multiple impact signals, which can reduce computational complexity and improve real-time performance. Third, the experimental results show that the proposed method maintains excellent feature recognition results compared to gate recurrent unit (GRU) and long short-term memory (LSTM) networks under all test datasets with varying impact speeds, while its metrics for computational complexity are 50% lower than those of the GRU and LSTM. Therefore, the proposed method is of great practical value for weak hardware application platforms that require the accurate identification of multiple dynamic impact signals in real time. [ABSTRACT FROM AUTHOR]

Data and reports indicate an increasing frequency and intensity of natural disasters worldwide. Buildings play a crucial role in disaster responses and damage assessments, aiding in planning rescue efforts and evaluating losses. Despite advances in applying deep learning to building extraction, challenges remain in handling complex natural disaster scenes and reducing reliance on labeled datasets. Recent advances in satellite video are opening a new avenue for efficient and accurate building extraction research. By thoroughly mining the characteristics of disaster video data, this work provides a new semantic segmentation model for accurate and efficient building extraction based on a limited number of training data, which consists of two parts: the prediction module and the automatic correction module. The prediction module, based on a base encoder–decoder structure, initially extracts buildings using a limited amount of training data that are obtained instantly. Then, the automatic correction module takes the output of the prediction module as input, constructs a criterion for identifying pixels with erroneous semantic information, and uses optical flow values to extract the accurate corresponding semantic information on the corrected frame. The experimental results demonstrate that the proposed method outperforms other methods in accuracy and computational complexity in complicated natural disaster scenes. [ABSTRACT FROM AUTHOR]

In this paper, the experimental results on microphone virtualization in realistic automotive scenarios are presented. A Temporal Convolutional Network (TCN) was designed in order to estimate the acoustic signal at the driver's ear positions based on the knowledge of monitoring microphone signals at different positions—a technique known as virtual microphone. An experimental setup was implemented on a popular B-segment car to acquire the acoustic field within the cabin while running on smooth asphalt at variable speeds. In order to test the potentiality of the TCN, microphone signals were recorded in two different scenarios, either with or without the front passenger. Our experimental results show that, when training is performed in both scenarios, the adopted TCN is able to robustly adapt to different conditions and guarantee a good average performance. Furthermore, an investigation on the parameters of the Neural Network (NN) that guarantee the sufficient accuracy of the estimation of the virtual microphone signals while maintaining a low computational complexity is presented. [ABSTRACT FROM AUTHOR]

Due to their high accuracy, excellent stability, minor size, and low cost, silicon piezoresistive pressure sensors are used to monitor downhole pressure under high-temperature, high-pressure conditions. However, due to silicon's temperature sensitivity, high and very varied downhole temperatures cause a significant bias in pressure measurement by the pressure sensor. The temperature coefficients differ from manufacturer to manufacturer and even vary from batch to batch within the same manufacturer. To ensure high accuracy and long-term stability for downhole pressure monitoring at high temperatures, this study proposes a temperature compensation method based on bilinear interpolation for piezoresistive pressure sensors under downhole high-temperature and high-pressure environments. A number of calibrations were performed with high-temperature co-calibration equipment to obtain the individual temperature characteristics of each sensor. Through the calibration, it was found that the output of the tested pressure measurement system is positively linear with pressure at the same temperatures and nearly negatively linear with temperature at the same pressures, which serves as the bias correction for the subsequent bilinear interpolation temperature compensation method. Based on this result, after least squares fitting and interpolating, a bilinear interpolation approach was introduced to compensate for temperature-induced pressure bias, which is easier to implement in a microcontroller (MCU). The test results show that the proposed method significantly improves the overall measurement accuracy of the tested sensor from 21.2% F.S. to 0.1% F.S. In addition, it reduces the MCU computational complexity of the compensation model, meeting the high accuracy demand for downhole pressure monitoring at high temperatures and pressures. [ABSTRACT FROM AUTHOR]

A novel and efficient modeling approach has been developed for simulating the responses of triaxial induction logging (TIL) in layered biaxial anisotropic (BA) formations. The core of this innovative technique lies in analytically calculating the primary fields within a homogeneous medium and approximating the scattered fields within layered formations. The former involves employing a two-level subtraction technique. Initially, the first-level subtraction entails altering the direction of the Fourier transform to mitigate the integral singularity of the spectral fields, particularly in high-angle and horizontal wells. Conversely, the second-level subtraction aims to further optimize integral convergence by creating an equivalent unbounded transverse isotropic (TI) formation and eliminating the corresponding spectral fields. With the two-level subtractions, the convergence of the spectral field has been enhanced by more than six orders of magnitude. Additionally, a strict recursive algorithm and approximation method are developed to compute the scattered fields in layered biaxial anisotropic media. The rigorous algorithm is based on a modified amplitude propagator matrix (MAPM) approach and serves as the benchmark for the approximation method. In contrast, the approximation method exploits the similarity between the spectral scattered field of the TI medium and the BA medium, establishing corresponding equivalent layered TI models for each magnetic component. Since the scattered field in TI models only involves a one-dimensional semi-infinite integral, the computational complexity is significantly reduced. Numerical simulation examples demonstrate that the new simulation method is at least two orders of magnitude faster than the current modeling approach while maintaining computational precision error within 0.5%. This significantly improved simulation efficiency provides a solid foundation for expediting the logging data processing. [ABSTRACT FROM AUTHOR]

The rapid development of unmanned aerial vehicle (UAV) technology has contributed to the increasing sophistication of UAV-based object-detection systems, which are now extensively utilized in civilian and military sectors. However, object detection from UAV images has numerous challenges, including significant variations in the object size, changing spatial configurations, and cluttered backgrounds with multiple interfering elements. To address these challenges, we propose SOD-YOLO, an innovative model based on the YOLOv8 model, to detect small objects in UAV images. The model integrates the receptive field convolutional block attention module (RFCBAM) in the backbone network to perform downsampling, improving feature extraction efficiency and mitigating the spatial information sparsity caused by downsampling. Additionally, we developed a novel neck architecture called the balanced spatial and semantic information fusion pyramid network (BSSI-FPN) designed for multi-scale feature fusion. The BSSI-FPN effectively balances spatial and semantic information across feature maps using three primary strategies: fully utilizing large-scale features, increasing the frequency of multi-scale feature fusion, and implementing dynamic upsampling. The experimental results on the VisDrone2019 dataset demonstrate that SOD-YOLO-s improves the mAP50 indicator by 3% compared to YOLOv8s while reducing the number of parameters and computational complexity by 84.2% and 30%, respectively. Compared to YOLOv8l, SOD-YOLO-l improves the mAP50 indicator by 7.7% and reduces the number of parameters by 59.6%. Compared to other existing methods, SODA-YOLO-l achieves the highest detection accuracy, demonstrating the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Aiming at the low observable target detection under sea clutter backgrounds, this paper emphasizes the exploration of distinguishable full-polarization features between target and sea clutter echoes. To overcome the shortcomings of the existing polarization feature-based methods, the full-polarization features of sea clutter are modeled and analyzed in detail by using Van Zyl polarization decomposition. Then, three polarimetric features (the relative surface scattering energy, the relative dihedral scattering energy and the relative diffuse scattering energy) are extracted from the fully polarimetric radar sea clutter echoes, which improve the feature differences between sea clutter and targets. And a tri-polarimetric feature detector with constant false alarm rate (CFAR) is constructed based on the fast convex hull learning algorithm. The experimental results on the real measured IPIX radar datasets prove that the proposed full-polarization feature detector obtains more competitive detection performance and lower computational complexity than the several existing feature-based detectors. [ABSTRACT FROM AUTHOR]

The transformer architecture is gradually gaining attention in remote sensing. Many algorithms related to this architecture have been proposed. However, the DEtection TRansformer (DETR) has been proposed as a new approach for implementing object detection tasks. It uses the transformer architecture for feature extraction, and its improved derivative models are uncommon in remote sensing object detection (RSOD). Hence, we selected the DETR with the improved deNoising anchor boxes (DINO) model as a foundation, upon which we have made improvements under the characteristics of remote sensing images (RSIs). Specifically, we proposed the skip-encoder (SE) module that can be applied to the encoder stage of the model and the skip-decoder (SD) module for the decoder stage. The SE module can enhance the model's ability to extract multiscale features. The SD module can reduce computational complexity and maintain the model performance. The experimental results on the NWPU VHR-10 and DIOR datasets demonstrate that the SE and SD modules can improve DINO for better learning small- and medium-sized targets in RSIs. We achieved a mean average precision of 94.8% on the NWPU VHR-10 dataset and 75.6% on the DIOR dataset. [ABSTRACT FROM AUTHOR]