Search Results (915)

Ocean Acoustic Tomography (OAT) is an important marine remote sensing technique used for inverting large-scale ocean environmental parameters, but traditional methods face challenges in computational complexity and environmental interference. This paper proposes a causal analysis-driven AI FOR SCIENCE method for high-precision and rapid inversion of oceanic hydrological parameters in complex underwater environments. Based on the open-source VTUAD (Vessel Type Underwater Acoustic Data) dataset, the method first utilizes a fine-tuned Paraformer (a fast and accurate parallel transformer) model for precise classification of sound source targets. Then, using structural causal models (SCM) and potential outcome frameworks, causal embedding vectors with physical significance are constructed. Finally, a cross-modal Transformer network is employed to fuse acoustic features, sound source priors, and environmental variables, enabling inversion of temperature and salinity in the Georgia Strait of Canada. Experimental results show that the method achieves accuracies of 97.77% and 95.52% for temperature and salinity inversion tasks, respectively, significantly outperforming traditional methods. Additionally, with GPU acceleration, the inference speed is improved by over sixfold, aimed at enabling real-time Ocean Acoustic Tomography (OAT) on edge computing platforms as smart hardware, thereby validating the method’s practicality. By incorporating causal inference and cross-modal data fusion, this study not only enhances inversion accuracy and model interpretability but also provides new insights for real-time applications of OAT. Full article

Based on the principle of symmetry, we propose a variable step-size FxLMS algorithm with double nonlinear functions cooperative coupling (DNVSS-FxLMS), aiming to optimize the contradiction between convergence rate and steady-state error in the active pressure pulsation control system of hydraulic systems. The algorithm innovatively couples two types of nonlinear mechanisms (rational-fractional and exponential-function-based), constructing a refined error-step mapping relationship to achieve a balance between rapid convergence and low steady-state error. Simulation experiments were conducted considering the complex time-varying operating environment of a simulation-based hydraulic system. The results demonstrate that, when the system undergoes unstable random changes, the DNVSS-FxLMS algorithm converges at least twice as fast as traditional and existing variable step size algorithms, while reducing steady-state error by 2–5 dB. The proposed DNVSS-FxLMS algorithm exhibits significant advantages in convergence rate, steady-state error reduction, and tracking capability, providing a highly efficient and robust solution for real-time active control of hydraulic system pressure pulsation under complex operating conditions. Full article

Wheat is one of the important grain crops, and spike counting is crucial for predicting spike yield. However, in complex farmland environments, the wheat body scale has huge differences, its color is highly similar to the background, and wheat ears often overlap with each other, which makes wheat ear detection work face a lot of challenges. At the same time, the increasing demand for high accuracy and fast response in wheat spike detection has led to the need for models to be lightweight function with reduced the hardware costs. Therefore, this study proposes a lightweight wheat ear detection model, CML-RTDETR, for efficient and accurate detection of wheat ears in real complex farmland environments. In the model construction, the lightweight network CSPDarknet is firstly introduced as the backbone network of CML-RTDETR to enhance the feature extraction efficiency. In addition, the FM module is cleverly introduced to modify the bottleneck layer in the C2f component, and hybrid feature extraction is realized by spatial and frequency domain splicing to enhance the feature extraction capability of wheat to be tested in complex scenes. Secondly, to improve the model’s detection capability for targets of different scales, a multi-scale feature enhancement pyramid (MFEP) is designed, consisting of GHSDConv, for efficiently obtaining low-level detail information and CSPDWOK for constructing a multi-scale semantic fusion structure. Finally, channel pruning based on Layer-Adaptive Magnitude Pruning (LAMP) scoring is performed to reduce model parameters and runtime memory. The experimental results on the GWHD2021 dataset show that the ${\mathrm{AP}}_{50}$ of CML-RTDETR reaches 90.5%, which is an improvement of 1.2% compared to the baseline RTDETR-R18 model. Meanwhile, the parameters and GFLOPs have been decreased to 11.03 M and 37.8 G, respectively, resulting in a reduction of 42% and 34%, respectively. Finally, the real-time frame rate reaches 73 fps, significantly achieving parameter simplification and speed improvement. Full article

The equivalent source method (ESM) is a core algorithm in integrated radiation-propagation acoustic field modeling. However, in challenging marine environments, including deep-sea and polar regions, where sound speed profiles exhibit strong vertical gradients, the ESM must increase waveguide stratification to maintain accuracy. This causes computational costs to scale exponentially with the number of layers, compromising efficiency and limiting applicability. To address this, this paper proposes a modal equivalent source (MES) model employing normal modes as basis functions instead of free-field Green’s functions. This model constructs a set of normal mode bases using full-depth hydroacoustic parameters, incorporating water column characteristics into the basis functions to eliminate waveguide stratification. This significantly reduces the computational matrix size of the ESM and computes acoustic fields in range-dependent waveguides using a single set of normal modes, resolving the dual limitations of inadequate precision and low efficiency in such environments. Concurrently, for the construction of basis functions, this paper also proposes a fast computation method for eigenvalues and eigenmodes in waveguide contexts based on phase functions and difference equations. Furthermore, coupling the MES method with the Finite Element Method (FEM) enables integrated computation of underwater target radiation and propagation fields. Multiple simulations demonstrate close agreement between the proposed model and reference results (errors < 4 dB). Under equivalent accuracy requirements, the proposed model reduces computation time to less than 1/25 of traditional ESM, achieving significant efficiency gains. Additionally, sea trial verification confirms model effectiveness, with mean correlation coefficients exceeding 0.9 and mean errors below 5 dB against experimental data. Full article

Object detection in aerial images can provide great support in traffic planning, national defense reconnaissance, hydrographic surveys, infrastructure construction, and other fields. Objects in aerial images are characterized by small pixel–area ratios, dense arrangements between objects, and arbitrary inclination angles. In response to these characteristics and problems, we improved the feature extraction network Inception-ResNet using the Fast Architecture Search (FAS) module and proposed a one-stage anchor-free rotation object detector. The structure of the object detector is simple and only consists of convolution layers, which reduces the number of model parameters. At the same time, the label sampling strategy in the training process is optimized to resolve the problem of insufficient sampling. Finally, a decoupled object detection head is used to separate the bounding box regression task from the object classification task. The experimental results show that the proposed method achieves mean average precision (mAP) of 82.6%, 79.5%, and 89.1% on the DOTA1.0, DOTA1.5, and HRSC2016 datasets, respectively, and the detection speed reaches 24.4 FPS, which can meet the needs of real-time detection. Full article

An electro-hydraulic servo loading system for a pavement mechanical properties test device was designed. The simulation analysis and test results showed that the PID control met the design requirements, but the output’s maximum error did not. Therefore, a fast terminal sliding mode control strategy with an extended state observer (ESO) was proposed. A tracking differentiator was constructed to obtain smooth differential signals from the input signals. The order of the system was reduced by considering the third and higher orders of the system as the total disturbance, and the states and the total disturbance of the system were estimated using the ESO. The fast terminal sliding mode control achieved fast convergence of the system within a limited time. The simulation results showed that the proposed control strategy improved the system accuracy and anti-disturbance ability, and system control performance was optimized. Full article

Sheep face recognition technology is critical in key areas such as individual sheep identification and behavior monitoring. Existing sheep face recognition models typically require high computational resources. When these models are deployed on mobile or embedded devices, problems such as reduced model recognition accuracy and increased recognition time arise. To address these problems, an improved Parameter Fusion Lightweight You Only Look Once (PFL-YOLO) sheep face recognition model based on YOLOv8n is proposed. In this study, the Efficient Hybrid Conv (EHConv) module is first integrated to enhance the extraction capability of the model for sheep face features. At the same time, the Residual C2f (RC2f) module is introduced to facilitate the effective fusion of multi-scale feature information and improve the information processing capability of the model; furthermore, the Efficient Spatial Pyramid Pooling Fast (ESPPF) module was used to fuse features of different scales. Finally, parameter fusion optimization work was carried out for the detection head, and the construction of the Parameter Fusion Detection (PFDetect) module was achieved, which significantly reduced the number of model parameters and computational complexity. The experimental results show that the PFL-YOLO model exhibits an excellent performance–efficiency balance in sheep face recognition tasks: mAP@50 and mAP@50:95 reach 99.5% and 87.4%, respectively, and the accuracy is close to or equal to the mainstream benchmark model. At the same time, the number of parameters is only 1.01 M, which is reduced by 45.1%, 83.7%, 66.6%, 71.4%, and 61.2% compared to YOLOv5n, YOLOv7-tiny, YOLOv8n, YOLOv9-t, and YOLO11n, respectively. The size of the model was compressed to 2.1 MB, which was reduced by 44.7%, 82.5%, 65%, 72%, and 59.6%, respectively, compared to similar lightweight models. The experimental results confirm that the PFL-YOLO model maintains high accuracy recognition performance while being lightweight and can provide a new solution for sheep face recognition models on resource-constrained devices. Full article

Resource-constrained Internet of Things (IoT) devices demand efficient and robust intrusion detection systems (IDSs) to counter evolving cyber threats. The traditional IDS models, however, struggle with high computational complexity and inadequate feature extraction, limiting their accuracy and generalizability in IoT environments. To address this, we propose FFT-RDNet, a lightweight IDS framework leveraging depthwise separable convolution and frequency-domain feature fusion. An ADASYN-Tomek Links hybrid strategy first addresses class imbalances. The core innovation of FFT-RDNet lies in its novel two-dimensional spatial feature modeling approach, realized through a dedicated dual-path feature embedding module. One branch extracts discriminative statistical features in the time domain, while the other branch transforms the data into the frequency domain via Fast Fourier Transform (FFT) to capture the essential energy distribution characteristics. These time–frequency domain features are fused to construct a two-dimensional feature space, which is then processed by a streamlined residual network using depthwise separable convolution. This network effectively captures complex periodic attack patterns with minimal computational overhead. Comprehensive evaluation on the NSL-KDD and CIC-IDS2018 datasets shows that FFT-RDNet outperforms state-of-the-art neural network IDSs across accuracy, precision, recall, and F1 score (improvements: 0.22–1%). Crucially, it achieves superior accuracy with a significantly reduced computational complexity, demonstrating high efficiency for resource-constrained IoT security deployments. Full article

Autonomous earthwork machinery is gaining traction as a means to boost productivity and safety on space-constrained urban sites, yet the fast-growing literature has not been fully integrated. To clarify current knowledge, we systematically searched Scopus and screened 597 records, retaining 157 peer-reviewed papers (2015–March 2025) that address autonomy, integrated control, or risk mitigation for excavators, bulldozers, and loaders. Descriptive statistics, VOSviewer mapping, and qualitative synthesis show the output rising rapidly and peaking at 30 papers in 2024, led by China, Korea, and the USA. Four tightly linked themes dominate: perception-driven machine autonomy, IoT-enabled integrated control systems, multi-sensor safety strategies, and the first demonstrations of fleet-level collaboration (e.g., coordinated excavator clusters and unmanned aerial vehicle and unmanned ground vehicle (UAV–UGV) site preparation). Advances include centimeter-scale path tracking, real-time vision-light detection and ranging (LiDAR) fusion and geofenced safety envelopes, but formal validation protocols and robust inter-machine communication remain open challenges. The review distils five research priorities, including adaptive perception and artificial intelligence (AI), digital-twin integration with building information modeling (BIM), cooperative multi-robot planning, rigorous safety assurance, and human–automation partnership that must be addressed to transform isolated prototypes into connected, self-optimizing fleets capable of delivering safer, faster, and more sustainable urban construction. Full article

The low-carbon transition of the global energy system is an urgent necessity to address climate change and meet growing energy demand. As a major source of energy consumption and emissions, buildings play a key role in this transition. This study systematically analyzes the flexible resources of building energy systems and vehicle-to-grid (V2G) interaction technologies, and mainly focuses on the regulation characteristics and coordination mechanisms of distributed energy supply (renewable energy and multi-energy cogeneration), energy storage (electric/thermal/cooling), and flexible loads (air conditioning and electric vehicles) within regional energy systems. The study reveals that distributed renewable energy and multi-energy cogeneration technologies form an integrated architecture through a complementary “output fluctuation mitigation–cascade energy supply” mechanism, enabling the coordinated optimization of building energy efficiency and grid regulation. Electricity and thermal energy storage serve as dual pillars of flexibility along the “fast response–economic storage” dimension. Air conditioning loads and electric vehicles (EVs) complement each other via thermodynamic regulation and Vehicle-to-Everything (V2X) technologies, constructing a dual-dimensional regulation mode in terms of both power and time. Ultimately, a dynamic balance system integrating sources, loads, and storage is established, driven by the spatiotemporal complementarity of multi-energy flows. This paper proposes an innovative framework that optimizes energy consumption and enhances grid stability by coordinating distributed renewable energy, energy storage, and flexible loads across multiple time scales. This approach offers a new perspective for achieving sustainable and flexible building energy systems. In addition, this paper explores the application of demand response policies in building energy systems, analyzing the role of policy incentives and market mechanisms in promoting building energy flexibility. Full article

Steel surface defect detection is a crucial step in ensuring industrial production quality. However, due to significant variations in scale and irregular geometric morphology of steel surface defects, existing detection algorithms show notable deficiencies in multi-scale feature representation and cross-layer multi-scale feature fusion efficiency. To address these challenges, this paper proposes an improved real-time steel surface defect detection model, NGD-YOLO, based on YOLOv5s, which achieves fast and high-precision defect detection under relatively low hardware conditions. Firstly, a lightweight and efficient Normalization-based Attention Module (NAM) is integrated into the C3 module to construct the C3NAM, enhancing multi-scale feature representation capabilities. Secondly, an efficient Gather–Distribute (GD) mechanism is introduced into the feature fusion component to build the GD-NAM network, thereby effectively reducing information loss during cross-layer multi-scale information fusion and adding a small target detection layer to enhance the detection performance of small defects. Finally, to mitigate the parameter increase caused by the GD-NAM network, a lightweight convolution module, DCConv, that integrates Efficient Channel Attention (ECA), is proposed and combined with the C3 module to construct the lightweight C3DC module. This approach improves detection speed and accuracy while reducing model parameters. Experimental results on the public NEU-DET dataset show that the proposed NGD-YOLO model achieves a detection accuracy of 79.2%, representing a 4.6% mAP improvement over the baseline YOLOv5s network with less than a quarter increase in parameters, and reaches 108.6 FPS, meeting the real-time monitoring requirements in industrial production environments. Full article

The decoding of motor imagery (MI) electroencephalogram (EEG) signals is crucial for motor control and rehabilitation. However, as feature extraction is the core component of the decoding process, traditional methods, often limited to single-feature domains or shallow time-frequency fusion, struggle to comprehensively capture the spatio-temporal-frequency characteristics of the signals, thereby limiting decoding accuracy. To address these limitations, this paper proposes a dual-branch neural network architecture with multi-domain feature fusion, the dual-branch spatio-temporal-frequency fusion convolutional network with Transformer (DB-STFFCNet). The DB-STFFCNet model consists of three modules: the spatiotemporal feature extraction module (STFE), the frequency feature extraction module (FFE), and the feature fusion and classification module. The STFE module employs a lightweight multi-dimensional attention network combined with a temporal Transformer encoder, capable of simultaneously modeling local fine-grained features and global spatiotemporal dependencies, effectively integrating spatiotemporal information and enhancing feature representation. The FFE module constructs a hierarchical feature refinement structure by leveraging the fast Fourier transform (FFT) and multi-scale frequency convolutions, while a frequency-domain Transformer encoder captures the global dependencies among frequency domain features, thus improving the model’s ability to represent key frequency information. Finally, the fusion module effectively consolidates the spatiotemporal and frequency features to achieve accurate classification. To evaluate the feasibility of the proposed method, experiments were conducted on the BCI Competition IV-2a and IV-2b public datasets, achieving accuracies of 83.13% and 89.54%, respectively, outperforming existing methods. This study provides a novel solution for joint time-frequency representation learning in EEG analysis. Full article

When a CMOS (Complementary Metal–Oxide–Semiconductor) imaging system operates at a high frame rate or a high line rate, the exposure time of the imaging system is limited, and the acquired image data will be dark, with a low signal-to-noise ratio and unsatisfactory sharpness. Therefore, in order to improve the visibility and signal-to-noise ratio of remote sensing images based on CMOS imaging systems, this paper proposes a low-light remote sensing image enhancement method and a corresponding ZYNQ (Zynq-7000 All Programmable SoC) design scheme called the BGIR (Bilateral-Guided Image Restoration) algorithm, which uses an improved multi-scale Retinex algorithm in the HSV (hue–saturation–value) color space. First, the RGB image is used to separate the original image’s H, S, and V components. Then, the V component is processed using the improved algorithm based on bilateral filtering. The image is then adjusted using the gamma correction algorithm to make preliminary adjustments to the brightness and contrast of the whole image, and the S component is processed using segmented linear enhancement to obtain the base layer. The algorithm is also deployed to ZYNQ using ARM + FPGA software synergy, reasonably allocating each algorithm module and accelerating the algorithm by using a lookup table and constructing a pipeline. The experimental results show that the proposed method improves processing speed by nearly 30 times while maintaining the recovery effect, which has the advantages of fast processing speed, miniaturization, embeddability, and portability. Following the end-to-end deployment, the processing speeds for resolutions of 640 × 480 and 1280 × 720 are shown to reach 80 fps and 30 fps, respectively, thereby satisfying the performance requirements of the imaging system. Full article

This paper presents new fast algorithms for the fourth type discrete Hartley transform (DHT-IV) for input data sequences of lengths from 3 to 8. Fast algorithms for small-sized trigonometric transforms can be used as building blocks for synthesizing algorithms for large-sized transforms. Additionally, they can be utilized to process small data blocks in various digital signal processing applications, thereby reducing overall system latency and computational complexity. The existing polynomial algebraic approach and the approach based on decomposing the transform matrix into cyclic convolution submatrices involve rather complicated housekeeping and a large number of additions. To avoid the noted drawback, this paper uses a structural approach to synthesize new algorithms. The starting point for constructing fast algorithms was to represent DHT-IV as a matrix–vector product. The next step was to bring the block structure of the DHT-IV matrix to one of the matrix patterns following the structural approach. In this case, if the block structure of the DHT-IV matrix did not match one of the existing patterns, its rows and columns were reordered, and, if necessary, the signs of some entries were changed. If this did not help, the DHT-IV matrix was represented as the sum of two or more matrices, and then each matrix was analyzed separately, if necessary, subjecting the matrices obtained by decomposition to the above transformations. As a result, the factorizations of matrix components were obtained, which led to a reduction in the arithmetic complexity of the developed algorithms. To illustrate the space–time structures of computational processes described by the developed algorithms, their data flow graphs are presented, which, if necessary, can be directly mapped onto the VLSI structure. The obtained DHT-IV algorithms can reduce the number of multiplications by an average of 75% compared with the direct calculation of matrix–vector products. However, the number of additions has increased by an average of 4%. Full article

Due to the limitations of the sowing machine performance and rice seed germination rates, missing seedlings inevitably occur after rice is sown in large fields. This phenomenon has a direct impact on the rice yield. In the field environment, the existing methods for detecting missing seedlings based on unmanned aerial vehicle (UAV) remote sensing images often have unsatisfactory effects. Therefore, to enable the fast and accurate detection of missing rice seedlings and facilitate subsequent reseeding, this study proposes a UAV remote-sensing-based method for detecting missing rice seedlings in large fields. The proposed method uses an improved PCERT-DETR model to detect rice seedlings and missing seedlings in UAV remote sensing images of large fields. The experimental results show that PCERT-DETR achieves an optimal performance on the self-constructed dataset, with an mean average precision (mAP) of 81.2%, precision (P) of 82.8%, recall (R) of 78.3%, and F₁-score (F₁) of 80.5%. The model’s parameter count is only 21.4 M and its FLOPs reach 66.6 G, meeting real-time detection requirements. Compared to the baseline network models, PCERT-DETR improves the P, R, F₁, and mAP by 15.0, 1.2, 8.5, and 6.8 percentage points, respectively. Furthermore, the performance evaluation experiments were carried out through ablation experiments, comparative detection model experiments and heat map visualization analysis, indicating that the model has a strong detection performance on the test set. The results confirm that the proposed model can accurately detect the number of missing rice seedlings. This study provides accurate information on the number of missing seedlings for subsequent reseeding operations, thus contributing to the improvement of precision farming practices. Full article