Search Results (982)

This paper presents a systematic comparative study of optical character recognition (OCR) architectures for Korean license plate recognition under identical detection conditions. Although recent automatic license plate recognition (ALPR) systems increasingly adopt Transformer-based decoders, it remains unclear whether performance differences arise primarily from sequence modeling strategies or from backbone feature representations. To address this issue, we employ a unified YOLOv12-based license plate detector and evaluate multiple OCR configurations, including a CNN with an Attention-LSTM decoder and a MobileNetV3 with a Transformer decoder. To ensure a fair comparison, a controlled ablation study is conducted in which the CNN backbone is fixed to ResNet-18 while varying only the sequence decoder. Experiments are performed on both static image datasets and tracking-based sequential datasets, assessing recognition accuracy, error characteristics, and processing speed across GPU and embedded platforms. The results demonstrate that the effectiveness of sequence decoders is highly dataset-dependent and strongly influenced by feature quality and region-of-interest (ROI) stability. Quantitative analysis further shows that tracking-induced error accumulation dominates OCR performance in sequential recognition scenarios. Moreover, Korean license plate–specific error patterns reveal failure modes not captured by generic OCR benchmarks. Finally, experiments on embedded platforms indicate that Transformer-based OCR models introduce significant computational and memory overhead, limiting their suitability for real-time deployment. These findings suggest that robust license plate recognition requires joint consideration of detection, tracking, and recognition rather than isolated optimization of OCR architectures. Full article

Euclidean distance transform (EDT) often suffers from high computational complexity and limited processing efficiency, especially when applied to large-scale images. To address these challenges, this paper proposes a GPU-based parallel EDT algorithm. The proposed approach first partitions the input image into multiple horizontal sub-blocks. For each sub-block, a row-wise recursive computation strategy is adopted to construct its Voronoi diagram in parallel, thereby reducing computational overhead by exploiting the strong structural similarity between the Voronoi diagrams of adjacent rows. Based on the Voronoi diagrams of all sub-blocks, the Euclidean distance from each pixel to the nearest background pixel is subsequently evaluated, completing the transform. Experimental results demonstrate that the proposed algorithm achieves up to a 52× speedup over traditional CPU-based EDT methods, leading to a substantial improvement in computational performance. Nevertheless, the scalability of the method is influenced by GPU memory capacity and the chosen sub-block partitioning strategy when processing extremely large images. Moreover, the core idea of leveraging inter-row Voronoi similarity to reduce redundant computation can be naturally extended to higher-dimensional exact EDT as well as approximate EDT variants. Full article

Recent developments in unmanned aerial vehicle (UAV) activity highlight the need for advanced electromagnetic spectrum monitoring systems that can detect drones operating near sensitive or restricted areas. Such systems can identify emissions from drones even under frequency-hopping conditions, providing an early warning system and enabling a timely response to protect critical infrastructure and ensure secure operations. In this context, the present work proposes the development of a high-performance multichannel broadband monitoring system with real-time analysis capabilities, designed on an SDR architecture based on USRP with three acquisition channels: two broadband (160 MHz and 80 MHz) and one narrowband (1 MHz) channel, for simultaneous, of extended spectrum segments, aligned with current requirements for analyzing emissions from drones in the 2.4 GHz and 5.8 GHz ISM bands. The processing system was configured to support cumulative bandwidths of over 200 MHz through a high-performance hardware platform (powerful CPU, fast storage, GPU acceleration) and fiber optic interconnection, ensuring stable and lossless transfer of large volumes of data. The proposed spectrum monitoring system proved to be extremely sensitive, flexible, and extensible, achieving a reception sensitivity of −130 dBm, thus exceeding the values commonly reported in the literature. Additionally, the parallel multichannel architecture facilitates real-time detection of signals from different frequency ranges and provides a foundation for advanced signal classification. Its reconfigurable design enables rapid adaptation to various signal types beyond unmanned aerial systems. Full article

AI-driven network security relies increasingly on Large Language Models (LLMs) to detect sophisticated threats; however, their deployment on resource-constrained edge devices is severely hindered by immense parameter scales. While unstructured pruning offers a theoretical reduction in model size, commodity Graphics Processing Unit (GPU) architectures fail to efficiently leverage element-wise sparsity due to the mismatch between fine-grained pruning patterns and the coarse-grained parallelism of Tensor Cores, leading to latency bottlenecks that compromise real-time analysis of high-volume security telemetry. To bridge this gap, we propose SPARTA (Sparse Parallel Architecture for Real-Time Threat Analysis), an algorithm–architecture co-design framework. Specifically, we integrate a hardware-based address remapping interface to enable flexible row-offset access. This mechanism facilitates a novel graph-based column vector merging strategy that aligns sparse data with Tensor Core parallelism, complemented by a pipelined execution scheme to mask decoding latencies. Evaluations on Llama2-7B and Llama2-13B benchmarks demonstrate that SPARTA achieves an average speedup of 2.35× compared to Flash-LLM, with peak speedups reaching 5.05×. These findings indicate that hardware-aware microarchitectural adaptations can effectively mitigate the penalties of unstructured sparsity, providing a viable pathway for efficient deployment in resource-constrained edge security. Full article

Navigation line extraction is essential for visual navigation in agricultural machinery, yet existing methods often perform poorly in complex environments due to challenges such as weed interference, broken crop rows, and leaf adhesion. To enhance the accuracy and robustness of crop row centerline identification, this study proposes an improved segmentation model based on SegNeXt with integrated adaptive region of interest (ROI) extraction for multi-growth-stage maize row perception. Improvements include constructing a Local module via pooling layers to refine contour features of seedling rows and enhance complementary information across feature maps. A multi-scale fusion attention (MFA) is also designed for adaptive weighted fusion during decoding, improving detail representation and generalization. Additionally, Focal Loss is introduced to mitigate background dominance and strengthen learning from sparse positive samples. An adaptive ROI extraction method was also developed to dynamically focus on navigable regions, thereby improving efficiency and localization accuracy. The outcomes revealed that the proposed model achieves a segmentation accuracy of 95.13% and an IoU of 93.86%. The experimental results show that the proposed algorithm achieves a processing speed of 27 frames per second (fps) on GPU and 16.8 fps on an embedded Jetson TX2 platform. This performance meets the real-time requirements for agricultural machinery operations. This study offers an efficient and reliable perception solution for vision-based navigation in maize fields. Full article

This paper presents an integrative methodology combining advanced network packet vectorization techniques, parallel processing with Dask, GPU-optimized machine learning models, and the Qdrant vector database. Our approach achieves a 99.9% detection rate for malicious traffic with only a 1% false-positive rate, setting new performance benchmarks for cybersecurity systems. The methodology establishes an average detection time limit not exceeding 10% of the total response time, maintaining high precision even for sophisticated attacks. The system processes 56 GB of PCAP files from Malware-Traffic-Analysis.net (2020–2024) through a five-stage pipeline: distributed packet processing, feature extraction, vectorization, vector database storage, and GPU-accelerated classification using XGBoost, Random Forest, and K-Nearest Neighbors models. Full article

This study aims to identify an effective neural network architecture for the task of semantic segmentation of the surface of beer wort at the stage of primary fermentation, using deep learning methodologies. Four contemporary architectures were evaluated and contrasted. The following networks are presented in both baseline and optimized forms: U-Net, DeepLabV3+, Fast-SCNN, and SegNet. The models were trained on a dataset of images depicting real beer surfaces at the primary fermentation stage. This was followed by the validation of the models using key metrics, including pixel classification accuracy, Mean Intersection over Union (mIoU), Dice Coefficient, inference time per image, and Graphics Processing Unit (GPU) resource utilization. Results indicate that the optimized U-Net achieved the optimal balance between performance and efficiency, attaining a validation accuracy of 88.85%, mIoU of 76.72%, and a Dice score of 86.71%. With an inference time of 49.5 milliseconds per image, coupled with minimal GPU utilization (18%), the model proves suitable for real-time deployment in production environments. Conversely, complex architectures, such as DeepLabV3+, did not yield the anticipated benefits, thereby underscoring the viability of utilizing compact models for highly specialized industrial tasks. This study establishes a novel quantitative metric for the assessment of fermentation. This is based on the characteristics of the foam surface and thus offers an objective alternative to traditional subjective inspections. The findings emphasize the potential of adapting optimized deep learning architectures to quality control tasks within the food industry, particularly in the brewing sector, and they pave the way for further integration into automated computer vision systems. Full article

Feature extraction and description are fundamental components of visual perception systems used in applications such as visual odometry, Simultaneous Localization and Mapping (SLAM), and autonomous navigation. In resource-constrained platforms, such as Unmanned Aerial Vehicles (UAVs), achieving real-time hardware acceleration on Field-Programmable Gate Arrays (FPGAs) is challenging. This work demonstrates an FPGA-based implementation of an adaptive ORB (Oriented FAST and Rotated BRIEF) feature extraction pipeline designed for high-throughput and energy-efficient embedded vision. The proposed architecture is a completely new design for the main algorithmic blocks of ORB, including the FAST (Features from Accelerated Segment Test) feature detector, Gaussian image filtering, moment computation, and descriptor generation. Adaptive mechanisms are introduced to dynamically adjust thresholds and filtering behavior, improving robustness under varying illumination conditions. The design is developed using a High-Level Synthesis (HLS) approach, where all processing modules are implemented as reusable hardware IP cores and integrated at the system level. The architecture is deployed and evaluated on two FPGA platforms, PYNQ-Z2 and KRIA KR260, and its performance is compared against CPU and GPU implementations using a dedicated C++ testbench based on OpenCV. Experimental results demonstrate significant improvements in throughput and energy efficiency while maintaining stable and scalable performance, making the proposed solution suitable for real-time embedded vision applications on UAVs and similar platforms. Notably, the FPGA implementation increases DSP utilization from 11% to 29% compared to the previous designs implemented by other researchers, effectively offloading computational tasks from general purpose logic (LUTs and FFs), reducing LUT usage by 6% and FF usage by 13%, while maintaining overall design stability, scalability, and acceptable thermal margins at 2.387 W. This work establishes a robust foundation for integrating the optimized ORB pipeline into larger drone systems and opens the door for future system-level enhancements. Full article

Smart Network Interface Cards (SmartNICs) represent a paradigm shift in system architecture by offloading packet processing and selected application logic from the host CPU to the network interface itself. This architectural evolution reduces end-to-end latency toward the physical limits of Ethernet while simultaneously decreasing CPU and memory bandwidth utilization. The current ecosystem comprises three principal categories of devices: (i) conventional fixed-function NICs augmented with limited offload capabilities; (ii) ASIC-based Data Processing Units (DPUs) that integrate multi-core processors and dedicated protocol accelerators; and (iii) FPGA-based SmartNIC shells—reconfigurable hardware frameworks that provide PCIe connectivity, DMA engines, Ethernet MAC interfaces, and control firmware, while exposing programmable logic regions for user-defined accelerators. This article provides a comparative survey of representative platforms from each category, with particular emphasis on open-source FPGA shells. It examines their architectural capabilities, programmability models, reconfiguration mechanisms, and support for GPU-centric peer-to-peer datapaths. Furthermore, it investigates the associated software stack, encompassing kernel drivers, user-space libraries, and control APIs. This study concludes by outlining open research challenges and future directions in RDMA-oriented data preprocessing and heterogeneous SmartNIC acceleration. Full article

With the rapid development of intelligent computing and artificial intelligence, there is an increasing demand for efficient, interpretable, and interactive frameworks for fine-grained text analysis. In the field of argument mining, existing approaches are often constrained by sentence-level processing, limited exploitation of key linguistic markers, and a lack of human–AI collaborative mechanisms, which restrict both recognition accuracy and computational efficiency. To address these challenges, this paper proposes KFF-Transformer, a computing-oriented human–AI collaborative framework for fine-grained argument element identification based on Toulmin’s model. The framework first employs an automatic key marker mining algorithm to expand a seed set of expert-labeled linguistic cues, significantly enhancing coverage and diversity. It then employs a lightweight deep learning architecture that combines BERT for contextual token encoding with a BiLSTM network enhanced by an attention mechanism to perform word-level classification of the six Toulmin elements. This approach leverages enriched key markers as critical features, enhancing both accuracy and interpretability. It should be noted that while our framework leverages BERT—a Transformer-based encoder—for contextual representation, the core sequence labeling module is based on BiLSTM and does not implement a standard Transformer block. Furthermore, a human-in-the-loop interaction mechanism is embedded to support real-time user correction and adaptive system refinement, improving robustness and practical usability. Experiments conducted on a dataset of 180 English argumentative essays demonstrate that KFF-Transformer identifies key markers in 1145 sentences and achieves an accuracy of 72.2% and an F1-score of 66.7%, outperforming a strong baseline by 3.7% and 2.8%, respectively. Moreover, the framework reduces processing time by 18.9% on CPU and achieves near-real-time performance of approximately 3.3 s on GPU. These results validate that KFF-Transformer effectively integrates linguistically grounded reasoning, efficient deep learning, and interactive design, providing a scalable and trustworthy solution for intelligent argument analysis in real-world educational applications. Full article

Accurate segmentation of 3D dental structures is essential for oral diagnosis, orthodontic planning, and digital dentistry. With the rapid advancement of 3D scanning and modeling technologies, high-resolution dental data have become increasingly common. However, existing approaches still struggle to process such high-resolution data efficiently. Current models often suffer from excessive parameter counts, slow inference, high computational overhead, and substantial GPU memory usage. These limitations compel many studies to downsample the input data to reduce training and inference costs—an operation that inevitably diminishes critical geometric details, blurs tooth boundaries, and compromises both fine-grained structural accuracy and model robustness. To address these challenges, this study proposes DiffusionNet++, an end-to-end segmentation framework capable of operating directly on raw high-resolution dental data. Building upon the standard DiffusionNet architecture, our method introduces a normal-enhanced multi-feature input strategy together with a lightweight SE channel-attention mechanism, enabling the model to effectively exploit local directional cues, curvature variations, and other higher-order geometric attributes while adaptively emphasizing discriminative feature channels. Experimental results demonstrate that the coordinates + normal feature configuration consistently delivers the best performance. DiffusionNet++ achieves substantial improvements in overall accuracy (OA), mean Intersection over Union (mIoU), and individual class IoU across all data types, while maintaining strong robustness and generalization on challenging cases, such as missing teeth and partially scanned data. Qualitative visualizations further corroborate these findings, showing superior boundary consistency, finer structural preservation, and enhanced recovery of incomplete regions. Overall, DiffusionNet++ offers an efficient, stable, and highly accurate solution for high-resolution 3D tooth segmentation, providing a powerful foundation for automated digital dentistry research and real-world clinical applications. Full article

During ion thruster operation, electromagnetic waves propagating through the plasma plume undergo absorption and refraction effects. This paper presents a graphics processing unit (GPU) parallel ray tracing (RT) algorithm for inhomogeneous media to analyze plasma plume-induced perturbations on the radiation characteristics of a satellite reflector antenna, substantially improving computational efficiency. This algorithm performs ray path tracing in the plume, with the vertex and central rays in each ray tube assigned to dedicated GPU threads. This enables the parallel computation of electromagnetic wave attenuation, phase, and polarization. By further applying aperture integration and the superposition principle, the influence of the plume on the far-field antenna radiation patterns is efficiently analyzed. Comparison with serial results validates the accuracy of the algorithm for plume calculation, achieving approximately 319 times speed-up for 586,928 ray tubes. Within the 2–5 GHz frequency range, the plume causes amplitude attenuation of less than 3 dB. This study provides an efficient solution for real-time analysis of plume-induced interference in satellite communications. Full article

Conventional sensing expends energy at three stages: powering dedicated sensors, transmitting measurements, and executing computationally intensive inference. Wireless sensing re-purposes WiFi channel state information (CSI) inherent in every packet, eliminating extra sensors and uplink traffic, though reliance on deep neural networks (DNNs) often trained and run on graphics processing units (GPUs) can negate these gains. This review highlights two core energy efficiency levers in CSI-based wireless sensing. First ambient CSI harvesting cuts power use by an order of magnitude compared to radar and active Internet of Things (IoT) sensors. Second, integrated sensing and communication (ISAC) embeds sensing functionality into existing WiFi links, thereby reducing device count, battery waste, and carbon impact. We review conventional handcrafted and accuracy-first methods to set the stage for surveying green learning strategies and lightweight learning techniques, including compact hybrid neural architectures, pruning, knowledge distillation, quantisation, and semi-supervised training that preserve accuracy while reducing model size and memory footprint. We also discuss hardware co-design from low-power microcontrollers to edge application-specific integrated circuits (ASICs) and WiFi firmware extensions that align computation with platform constraints. Finally, we identify open challenges in domain-robust compression, multi-antenna calibration, energy-proportionate model scaling, and standardised joules per inference metrics. Our aim is a practical battery-friendly wireless sensing stack ready for smart home and 6G era deployments. Full article

Metal additive manufacturing (AM) generates complex microstructures through extreme thermal gradients and rapid solidification, critically influencing mechanical performance and industrial qualification. This review synthesizes recent advances in cellular automata (CA) and phase-field (PF) modeling to predict grain-scale microstructure evolution during AM. CA methods provide computational efficiency, enabling large-domain simulations and excelling in texture prediction and multi-layer builds. PF approaches deliver superior thermodynamic fidelity for interface dynamics, solute partitioning, and nonequilibrium rapid solidification through CALPHAD coupling. Hybrid CA–PF frameworks strategically balance efficiency and accuracy by allocating PF to solidification fronts and CA to bulk grain competition. Recent algorithmic innovations—discrete event-inspired CA, GPU acceleration, and machine learning—extend scalability while maintaining predictive capability. Validated applications across Ni-based superalloys, Ti-6Al-4V, tool steels, and Al alloys demonstrate robust process–microstructure–property predictions through EBSD and mechanical testing. Persistent challenges include computational scalability for full-scale components, standardized calibration protocols, limited in situ validation, and incomplete multi-physics coupling. Emerging solutions leverage physics-informed machine learning, digital twin architectures, and open-source platforms to enable predictive microstructure control for first-time-right manufacturing in aerospace, biomedical, and energy applications. Full article

Geophysical forward modeling serves as a fundamental theoretical approach for characterizing subsurface structures and material properties, essentially involving the computation of gravity responses at surface or spatial observation points based on a predefined density distribution. With the rapid development of data-driven techniques such as deep learning in geophysical inversion, forward algorithms are facing increasing demands in terms of computational scale, observable types, and efficiency. To address these challenges, this study develops an efficient forward modeling method based on voxel discretization, the enabling rapid calculation of gravity anomalies and radial gravity gradients on multiple observational surfaces. Leveraging the parallel computing capabilities of graphics processing units (GPU), together with tensor acceleration, Compute Unified Device Architecture (CUDA) execution, and Just-in-time (JIT) compilation strategies, the method achieves high efficiency and automation in the forward computation process. Numerical experiments conducted on several typical theoretical models demonstrate the convergence and stability of the calculated results, indicating that the proposed method significantly reduces computation time while maintaining accuracy, thus being well-suited for large-scale 3D modeling and fast batch simulation tasks. This research can efficiently generate forward datasets with multi-view and multi-metric characteristics, providing solid data support and a scalable computational platform for deep-learning-based geophysical inversion studies. Full article