Search Results (653)

With the continuous advancement of 3D perception technology, point cloud data has found increasingly widespread application. However, the presence of holes in point cloud data caused by device limitations and environmental interference severely restricts algorithmic performance, making point cloud completion a research topic of high interest. This study observes that most existing mainstream point cloud completion methods primarily focus on capturing global features, while often underrepresenting local structural details. Moreover, the generation process of complete point clouds lacks effective control over fine-grained features, leading to insufficient detail in the completed outputs and reduced data integrity. To address these issues, we propose a Set Combination Multi-Layer Perceptron (SCMP) module that enables the simultaneous extraction of both local and global features, thereby reducing the loss of local detail information. In addition, we introduce the Squeeze Excitation Pooling Network (SEP-Net) module, an informative channel attention mechanism capable of adaptively identifying and enhancing critical channel features, thus improving the overall feature representation capability. Based on these modules, we further design a novel Feature Fusion Point Fractal Network (FFPF-Net), which fuses multi-dimensional point cloud features to enhance representation capacity and progressively refines the missing regions to generate a more complete point cloud. Extensive experiments conducted on the ShapeNet-Part and MVP datasets compared to L-GAN and PCN showed average prediction error improvements of 1.3 and 1.4, respectively. The average completion errors on the ShapeNet-Part and MVP datasets are 0.783 and 0.824, highlighting the improved fine-detail reconstruction capability of our network. These results indicate that the proposed method effectively enhances point cloud completion performance and can further promote the practical application of point cloud data in various real-world scenarios. Full article

In massive multiple-input and multiple-output (mMIMO) systems, superimposed channel state information (CSI) feedback is developed to improve the occupation of uplink bandwidth resources. Nevertheless, the interference from this superimposed mode degrades the recovery performance of both downlink CSI and uplink data sequences. Although machine learning (ML)-based methods effectively mitigate superimposed interference by leveraging the multi-domain features of downlink CSI, the complex interactions among network model parameters cause a significant burden on system resources. To address these issues, inspired by sensing-assisted communication, we propose a novel superimposed CSI feedback method assisted by inactive sensing information that previously existed but was not utilized at the base station (BS). To the best of our knowledge, this is the first time that inactive sensing information is utilized to enhance superimposed CSI feedback. In this method, a new type of modal data, different from communication data, is developed to aid in interference suppression without requiring additional hardware at the BS. Specifically, the proposed method utilizes location, speed, and path information extracted from sensing devices to derive prior information. Then, based on the derived prior information, denoising processing is applied to both the delay and Doppler dimensions of downlink CSI in the delay—Doppler (DD) domain, significantly enhancing the recovery accuracy. Simulation results demonstrate the performance improvement of downlink CSI and uplink data sequences when compared to both classic and novel superimposed CSI feedback methods. Moreover, against parameter variations, simulation results also validate the robustness of the proposed method. Full article

Entropy-based analyses have emerged as a powerful tool for quantifying the complexity, regularity, and information content of complex biological signals, such as electroencephalography (EEG). In this regard, EEG-based lie detection offers the advantage of directly providing more objective and less susceptible-to-manipulation results compared to traditional polygraph methods. To this end, this study proposes a novel multi-scale entropy approach by fusing fuzzy entropy (FE), time-shifted multi-scale fuzzy entropy (TSMFE), and hierarchical multi-band fuzzy entropy (HMFE), which enables the multidimensional characterization of EEG signals. Subsequently, using machine learning classifiers, the fused feature vector is applied to lie detection, with a focus on channel selection to investigate distinguished neural signatures across brain regions. Experiments utilize a publicly benchmarked LieWaves dataset, and two parts are performed. One is a subject-dependent experiment to identify representative channels for lie detection. Another is a cross-subject experiment to assess the generalizability of the proposed approach. In the subject-dependent experiment, linear discriminant analysis (LDA) achieves impressive accuracies of 82.74% under leave-one-out cross-validation (LOOCV) and 82.00% under 10-fold cross-validation. The cross-subject experiment yields an accuracy of 64.07% using a radial basis function (RBF) kernel support vector machine (SVM) under leave-one-subject-out cross-validation (LOSOCV). Furthermore, regarding the channel selection results, PZ (parietal midline) and T7 (left temporal) are considered the representative channels for lie detection, as they exhibit the most prominent occurrences among subjects. These findings demonstrate that the PZ and T7 play vital roles in the cognitive processes associated with lying, offering a solution for designing portable EEG-based lie detection devices with fewer channels, which also provides insights into neural dynamics by analyzing variations in multi-scale entropy. Full article

This study provides an in-depth analysis of the learning dynamics of multichannel optical neurons based on spatial solitons generated in lithium niobate crystals. Single-node and multi-node configurations with different topological complexities (3 × 3, 4 × 4, and 5 × 5) were compared, assessing how the number of channels, geometry, and optical parameters affect the speed and efficiency of learning. The simulations indicate that single-node neurons achieve the desired imbalance more rapidly and with lower energy expenditure, whereas multi-node structures require higher intensities and longer timescales, yet yield a greater variety of responses, more accurately reproducing the functional diversity of biological neural tissues. The results highlight how the plasticity of these devices can be entirely modulated through optical parameters, paving the way for fully optical photonic neuromorphic networks in which memory and computation are co-localized, with potential applications in on-chip learning, adaptive routing, and distributed decision-making. Full article

Accurate and efficient detection of underwater targets in sonar imagery is critical for applications such as marine exploration, infrastructure inspection, and autonomous navigation. However, sonar-based object detection remains challenging due to low resolution, high noise, cluttered backgrounds, and the scarcity of annotated data. To address these issues, we propose UWS-YOLO, a novel detection framework specifically designed for underwater sonar images. The model integrates three key innovations: (1) a C2F-Ortho module that enhances multi-scale feature representation through orthogonal channel attention, improving sensitivity to small and low-contrast targets; (2) a DySnConv module that employs Dynamic Snake Convolution to adaptively capture elongated and irregular structures such as pipelines and cables; and (3) a cross-modal transfer learning strategy that pre-trains on large-scale optical underwater imagery before fine-tuning on sonar data, effectively mitigating overfitting and bridging the modality gap. Extensive evaluations on real-world sonar datasets demonstrate that UWS-YOLO achieves a mAP@0.5 of 87.1%, outperforming the YOLOv8n baseline by 3.5% and seven state-of-the-art detectors in accuracy while maintaining real-time performance at 158 FPS with only 8.8 GFLOPs. The framework exhibits strong generalization across datasets, robustness to noise, and computational efficiency on embedded devices, confirming its suitability for deployment in resource-constrained underwater environments. Full article

Leaves are the key organs in photosynthesis and nutrient production, and leaf counting is an important indicator of banana plant health and growth rate. However, in complex orchard environments, leaves often overlap, the background is cluttered, and illumination varies, making accurate segmentation and detection challenging. To address these issues, we propose a lightweight banana leaf detection and counting method deployable on embedded devices, which integrates a space–depth-collaborative reasoning strategy with multi-scale feature enhancement to achieve efficient and precise leaf identification and counting. For complex background interference and occlusion, we design a multi-scale attention guided feature enhancement mechanism that employs a Mixed Local Channel Attention (MLCA) module and a Self-Ensembling Attention Mechanism (SEAM) to strengthen local salient feature representation, suppress background noise, and improve discriminability under occlusion. To mitigate feature drift caused by environmental changes, we introduce a task-aware dynamic scale adaptive detection head (DyHead) combined with multi-rate depthwise separable dilated convolutions (DWR_Conv) to enhance multi-scale contextual awareness and adaptive feature recognition. Furthermore, to tackle instance differentiation and counting under occlusion and overlap, we develop a detection-guided space–depth position modeling method that, based on object detection, effectively models the distribution of occluded instances through space–depth feature description, outlier removal, and adaptive clustering analysis. Experimental results demonstrate that our YOLOv8n MDSD model outperforms the baseline by 2.08% in mAP50-95, and achieves a mean absolute error (MAE) of 0.67 and a root mean square error (RMSE) of 1.01 in leaf counting, exhibiting excellent accuracy and robustness for automated banana leaf statistics. Full article

Healthcare Internet of Things (H-IoT) systems demand ultra-reliable and low-latency communication (URLLC) to support critical functions such as remote monitoring, emergency response, and real-time diagnostics. However, spectrum scarcity and heterogeneous traffic patterns pose major challenges for centralized scheduling in dense H-IoT deployments. This paper proposed a multi-agent reinforcement learning (MARL) framework for dynamic, priority-aware spectrum management (PASM), where cooperative MARL agents jointly optimize throughput, latency, energy efficiency, fairness, and blocking probability under varying traffic and channel conditions. Six learning strategies are developed and compared, including Q-Learning, Double Q-Learning, Deep Q-Network (DQN), Actor–Critic, Dueling DQN, and Proximal Policy Optimization (PPO), within a simulated H-IoT environment that captures heterogeneous traffic, device priorities, and realistic URLLC constraints. A comprehensive simulation study across scalable scenarios ranging from 3 to 50 devices demonstrated that PPO consistently outperforms all baselines, improving mean throughput by $6.2\%$, reducing 95th-percentile delay by $11.5\%$, increasing energy efficiency by $11.9\%$, lowering blocking probability by $33.3\%$, and accelerating convergence by $75.8\%$ compared to the strongest non-PPO baseline. These findings establish PPO as a robust and scalable solution for QoS-compliant spectrum management in dense H-IoT environments, while Dueling DQN emerges as a competitive deep RL alternative. Full article

Drowning constitutes the leading cause of injury-related fatalities among adolescents. In swimming pool environments, traditional manual surveillance exhibits limitations, while existing technologies suffer from poor adaptability of wearable devices. Vision models based on YOLO still face challenges in edge deployment efficiency, robustness in complex water conditions, and multi-scale object detection. To address these issues, we propose YOLO11-LiB, a drowning object detection model based on YOLO11n, featuring three key enhancements. First, we design the Lightweight Feature Extraction Module (LGCBlock), which integrates the Lightweight Attention Encoding Block (LAE) and effectively combines Ghost Convolution (GhostConv) with dynamic convolution (DynamicConv). This optimizes the downsampling structure and the C3k2 module in the YOLO11n backbone network, significantly reducing model parameters and computational complexity. Second, we introduce the Cross-Channel Position-aware Spatial Attention Inverted Residual with Spatial–Channel Separate Attention module (C2PSAiSCSA) into the backbone. This module embeds the Spatial–Channel Separate Attention (SCSA) mechanism within the Inverted Residual Mobile Block (iRMB) framework, enabling more comprehensive and efficient feature extraction. Finally, we redesign the neck structure as the Bidirectional Feature Fusion Network (BiFF-Net), which integrates the Bidirectional Feature Pyramid Network (BiFPN) and Frequency-Aware Feature Fusion (FreqFusion). The enhanced YOLO11-LiB model was validated against mainstream algorithms through comparative experiments, and ablation studies were conducted. Experimental results demonstrate that YOLO11-LiB achieves a drowning class mean average precision (DmAP50) of 94.1%, with merely 2.02 M parameters and a model size of 4.25 MB. This represents an effective balance between accuracy and efficiency, providing a high-performance solution for real-time drowning detection in swimming pool scenarios. Full article

Internet of Things (IoT) applications are rapidly adopting low-power wide-area network (LPWAN) technology due to its ability to provide broad coverage for a range of battery-powered devices. LoRaWAN has become the most widely used LPWAN solution due to its physical layer (PHY) design and regulatory advantages. Because LoRaWAN has a broad communication range, the coverage of the gateways might overlap. In LoRa technology, packets can be received concurrently by multiple gateways. Subsequently, the network server selects the packet with the highest receiver strength signal indicator (RSSI). However, this method can lead to the exhaustion of channel availability on the gateways. The optimisation of configuration parameters to reduce collisions and enhance network throughput in multi-gateway LoRaWAN remains an unresolved challenge. This paper introduces a novel low-complexity model for ZBMG-LoRa, mitigates the collisions using channel utilisiation, and categorises nodes into distinct groups based on their respective gateways. This categorisation allows for the implementation of optimal settings for each node’s subzone, thereby facilitating effective communication and addressing the identified issue. By deriving key performance metrics (e.g., network throughput, energy efficiency, and probability of effective delivery) from configuration parameters and network size, communication reliability is maintained. Optimal configurations for transmission power and spreading factor are derived by our method for all nodes in LoRaWAN networks with multiple gateways. In comparison to adaptive data rate (ADR) and other related state-of-the-art algorithms, the findings demonstrate that the novel approach achieves higher packet delivery ratio and better energy efficiency. Full article

To ensure the safety and quality of grain during storage requires distributed monitoring of temperature and relative humidity within the bulk material, where hundreds of sensors may be needed. Conventional multi-channel systems are often constrained by the limited number of sensors connectable to a single acquisition unit, high hardware cost, and poor scalability. To address these challenges, this study proposes a novel design method for a multi-channel temperature and relative humidity acquisition system (MTRHAS). The system integrates sequential sampling control and a time-division multiplexing mechanism, enabling efficient data acquisition from multiple sensors while reducing hardware requirements and cost. This system employs sequential sampling control using a single complex programmable logic device (CPLD), and uses multiple CPLDs for multi-channel sensor expansion with a shared address and data bus for communication with a microcontroller unit (MCU). A prototype was developed using two CPLDs and one MCU, achieving data collection from 80 sensors. To validate the approach, a simulated grain silo experiment was conducted, with nine sensors deployed to monitor temperature and relative humidity during aeration. Calibration ensured sensor accuracy, and real-time monitoring results revealed that the system effectively captured spatial and temporal variation patterns of intergranular air conditions. Compared with conventional designs, the proposed system shortens the sampling cycle, decreases the number of acquisition units required, and enhances scalability through the shared bus architecture. These findings demonstrate that the MTRHAS provides an efficient and practical solution for large-scale monitoring of grain storage environments. Full article

Polarization information is essential for material identification, stress mapping, biological imaging, and robust vision under strong illumination, yet conventional approaches rely on external polarization optics and active biasing, which are bulky, alignment-sensitive, and power-hungry. A more desirable route is to encode polarization at the pixel level and read it out at zero bias, enabling compact, low-noise, and polarization imaging. Low-symmetry layered semiconductors provide persistent in-plane anisotropy as a materials basis for polarization selectivity. Here, we construct an eight-terminal radial ‘star-shaped’ Au/ReS₂ metal-semiconductor junction array pixel that operates in a genuine photovoltaic mode under zero external bias based on the photothermoelectric effect. Based on this, electrical summation of phase-matched multi-junction channels increases the signal amplitude approximately linearly without sacrificing the two-lobed modulation depth, achieving ‘gain by stacking’ without external amplification. The device exhibits millisecond-scale transient response and robust cycling stability and, as a minimal pixel unit, realizes polarization-resolved imaging and pattern recognition. Treating linear combinations of channels as operators in the polarization domain, these results provide a general pixel-level foundation for compact, zero-bias, and scalable polarization cameras and on-pixel computational sensing. Full article

The growing challenges in healthcare services, such as hospital congestion and a persistent shortage of medical personnel, significantly impede effective service delivery. This particularly presents significant challenges for the continuous monitoring of patients with chronic diseases. In comparison, Internet of Things (IoT) based telemonitoring systems offer a promising solution to alleviate these challenges. However, transmitting sensitive and confidential patient health data requires a strong focus on end-to-end security. This includes securing sensitive data within the patient’s home network, during internet transmission, at the endpoint system, and managing sensitive data. In this study, we propose a secure and scalable architecture for a remote health monitoring system that integrates telemedicine technology with the IoT. The proposed solution included a portable remote health monitoring device, an IoT Gateway appliance (IoT GW) in the patient’s home, and an IoT Application Gateway Endpoint (App GW Endpoint) on a cloud infrastructure. A secure communication channel was established by implementing a multi-layered security protocol stack that uses HTTPS over Quick UDP Internet Connection (QUIC), with a focus on optimal security and compatibility, prioritizing cipher suites for data confidentiality and device authentication. The cloud architecture is designed based on the Well-Architected Framework principles to ensure security, high availability, and scalability. Our study shows that storing patient health information is reliable and efficient. Furthermore, the results for processing and transmission times clearly demonstrate that the additional encryption mechanisms have a negligible effect on data transmission latency, while significantly improving security. Full article

The inspection of the appearance quality of tea leaves is vital for market classification and value assessment within the tea industry. Nevertheless, many existing detection approaches rely on sophisticated model architectures, which hinder their practical use on devices with limited computational resources. This study proposes a lightweight object detection network, TeaAppearanceLiteNet, tailored for tea leaf appearance analysis. A novel C3k2_PartialConv module is introduced to significantly reduce computational redundancy while maintaining effective feature extraction. The CBMA_MSCA attention mechanism is incorporated to enable the multi-scale modeling of channel attention, enhancing the perception accuracy of features at various scales. By incorporating the Detect_PinwheelShapedConv head, the spatial representation power of the network is significantly improved. In addition, the MPDIoU_ShapeIoU loss is formulated to enhance the correspondence between predicted and ground-truth bounding boxes across multiple dimensions—covering spatial location, geometric shape, and scale—which contributes to a more stable regression and higher detection accuracy. Experimental results demonstrate that, compared to baseline methods, TeaAppearanceLiteNet achieves a 12.27% improvement in accuracy, reaching a mAP@0.5 of 84.06% with an inference speed of 157.81 FPS. The parameter count is only 1.83% of traditional models. The compact and high-efficiency design of TeaAppearanceLiteNet enables its deployment on mobile and edge devices, thereby supporting the digitalization and intelligent upgrading of the tea industry under the framework of smart agriculture. Full article

Individual cow identification is a prerequisite for automatically monitoring behavior patterns, health status, and growth data of each cow, and can provide the assistance in selecting excellent cow individuals for breeding. Despite high recognition accuracy, traditional implantable electronic devices such as RFID (i.e., Radio Frequency Identification) can cause some degree of harm or stress reactions to cows. Image-based methods are widely used due to their non-invasive advantages, but these methods have poor adaptability to different environments and target size, and low detection accuracy in complex scenes. To solve these issues, this study designs a Dy_Conv (i.e., dynamic convolution) module and innovatively constructs a Dynamic_Bottleneck module based on the Dy_Conv and S2Attention (Sparse-shift Attention) mechanism. On this basis, we replaces the first and fourth bottleneck layers of Resnet50 with the Dynamic_Bottleneck to achieve accurate extraction of local features and global information of cows. Furthermore, the QAConv (i.e., query adaptive convolution) module is introduced into the front end of the backbone network, and can adjust the parameters and sizes of convolution kernels to adapt to the scale changes in cow targets and input images. At the same time, NAM (i.e., normalization-based attention module) attention is embedded into the backend of the network to achieve the feature fusion in the channels and spatial dimensions, which contributes to better distinguish visually similar individual cows. The experiments are conducted on the public datasets collected from different cowsheds. The experimental results showed that the Rank-1, Rank-5, and mAP metrics reached 96.8%, 98.9%, and 95.3%, respectively. Therefore, the proposed model can effectively capture and integrate multi-scale features of cow body appearance, enhancing the accuracy of individual cow identification in complex scenes. Full article

Plant disease segmentation has achieved significant progress with the help of artificial intelligence. However, deploying high-accuracy segmentation models in resource-limited settings faces three key challenges, as follows: (A) Traditional dense attention mechanisms incur quadratic computational complexity growth (O($n^{2}d$)), rendering them ill-suited for low-power hardware. (B) Naturally sparse spatial distributions and large-scale variations in the lesions on leaves necessitate models that concurrently capture long-range dependencies and local details. (C) Complex backgrounds and variable lighting in field images often induce segmentation errors. To address these challenges, we propose Sparse-MoE-SAM, an efficient framework based on an enhanced Segment Anything Model (SAM). This deep learning framework integrates sparse attention mechanisms with a two-stage mixture of experts (MoE) decoder. The sparse attention dynamically activates key channels aligned with lesion sparsity patterns, reducing self-attention complexity while preserving long-range context. Stage 1 of the MoE decoder performs coarse-grained boundary localization; Stage 2 achieves fine-grained segmentation by leveraging specialized experts within the MoE, significantly enhancing edge discrimination accuracy. The expert repository—comprising standard convolutions, dilated convolutions, and depthwise separable convolutions—dynamically routes features through optimized processing paths based on input texture and lesion morphology. This enables robust segmentation across diverse leaf textures and plant developmental stages. Further, we design a sparse attention-enhanced Atrous Spatial Pyramid Pooling (ASPP) module to capture multi-scale contexts for both extensive lesions and small spots. Evaluations on three heterogeneous datasets (PlantVillage Extended, CVPPP, and our self-collected field images) show that Sparse-MoE-SAM achieves a mean Intersection-over-Union (mIoU) of 94.2%—surpassing standard SAM by 2.5 percentage points—while reducing computational costs by 23.7% compared to the original SAM baseline. The model also demonstrates balanced performance across disease classes and enhanced hardware compatibility. Our work validates that integrating sparse attention with MoE mechanisms sustains accuracy while drastically lowering computational demands, enabling the scalable deployment of plant disease segmentation models on mobile and edge devices. Full article