Search Results (356)

Floods remain among the most devastating natural disasters globally, disproportionately impacting low-resource regions where real-time flood forecasting is constrained by limited computational infrastructure and the scarcity of fine-resolution predictive models. Although state-of-the-art global climate models achieve high predictive accuracy, their scale and computational complexity restrict their applicability in localized and resource-constrained settings. This study proposes a deep climate model distillation framework that transfers knowledge from a high-capacity Fourier Neural Operator (FNO)-based global climate model inspired by FourCastNet into lightweight, regionally adaptive student networks suitable for edge deployment. The framework combines climate variables, satellite observations, and hydrological measurements to improve localized flood prediction. Knowledge transfer is achieved through a multi-objective distillation strategy that combines supervised learning, soft-target alignment, and intermediate feature matching. Experimental evaluation across multiple flood-prone regions in Sub-Saharan Africa and South Asia shows that the distilled student model achieves an average classification accuracy of 0.89, an AUC of 0.91, and an F1-score of 0.88, retaining approximately 96.7% of the teacher model’s predictive performance. In continuous discharge estimation, the model attains a mean absolute error of 0.17, RMSE of 0.24, and an R² score of 0.85. The proposed distillation approach yields an 8× reduction in inference latency and over a 20× reduction in model size, enabling real-time execution on low-power edge devices such as the Raspberry Pi 4 and NVIDIA Jetson Nano. The student model further demonstrates robust regional and temporal generalization, with limited performance degradation in unseen geographic areas and during extreme flood years. Full article

Stereo vision is essential for passive three-dimensional perception in resource-constrained applications that require low power consumption, predictable latency, and explainable geometry. Although deep learning architectures dominate recent benchmarks, the classical block-matching pipeline remains a foundational approach. Optimizing this pipeline involves navigating complex trade-offs among matching robustness, map density, and computational efficiency. This study systematically surveys and physically validates the classical stereo framework. After revisiting geometric first principles, three matching costs (SAD, NCC, ZNCC) are benchmarked alongside Sobel preprocessing and structural refinements, with subsequent validation using a calibrated consumer webcam rig. Middlebury benchmarks (2001–2021) indicate that while SAD fails under complex radiometric distortion, NCC consistently achieves superior quantitative metrics, incurring only a 1.2-fold computational overhead. Extending the disparity search range improves foreground localization, while block size imposes a trade-off between resolving the aperture problem and preserving fine geometric detail. To bridge theoretical analysis and practical deployment, the pipeline is validated using a custom-calibrated consumer stereo rig. The optimized Sobel-NCC architecture is then evaluated for real-time edge deployment on constrained hardware (NVIDIA Jetson Nano) and narrow-baseline sensors (OAK-D SR) in the context of agricultural robotic manipulation. By prioritizing metric precision over dense prediction, the classical pipeline reconstructs target surfaces with approximately 1 cm depth accuracy at 21 frames per second. These results demonstrate that optimized local algorithms offer deterministic and reliable geometric foundations for real-time edge-computed robotics. Although neural networks are essential for dense reconstructions in ill-posed regions, the foundational principles established here remain indispensable for advanced stereo vision system deployment. Full article

In UAV-based remote sensing, accurate and efficient image mosaicing is crucial for achieving real-time monitoring. Traditional cloud-centric processing paradigms, however, face core scientific challenges such as high latency, bandwidth bottlenecks, and limited autonomy, making them inadequate for dynamic, real-time scenarios. To address these issues, this paper proposes an edge-computing-enabled UAV image mosaicing system. The system consists of a UAV remote sensing platform and an edge computing terminal, with the core being our novel B-SIFT-ILS algorithm. The algorithm first uses geographic coordinates for unified registration, constructs a Gaussian scale space for multi-resolution representation, and then precisely locates extrema in the Difference of Gaussian (DoG) space using a 3D quadratic function. A BANSAC algorithm is subsequently employed to refine feature points and extract stable SIFT features, and finally, Iterative Least Squares (ILS) are used to achieve seamless mosaicing. Experimental results demonstrate that, compared with classical RANSAC, the proposed method achieves superior feature sampling accuracy (rotation: 0.879, translation: 0.877) and lower latency. The ILS-based smoothing stage effectively eliminates noise and ghosting without introducing gradient reversal, performing comparably to deep learning methods while significantly outperforming direct averaging and Gaussian approaches. On the NVIDIA Jetson Orin NX edge terminal, a single processing instance requires only 1124 ms, highlighting its strong potential for real-time, low-latency, and autonomous mosaicing tasks. Future research will focus on extending the approach to non-planar terrains and implementing adaptive parameter tuning for the BANSAC algorithm. Full article

As a measure to sustain crops, the presence of irrigation, man-made reservoirs has become very common in regions affected by prolonged periods of low rainfall. Although these reservoirs must be provided with minimum safety facilities, it is also very common that animals or, to a much lesser extent, the people in charge of their maintenance, fall into the reservoir. The reservoirs then become, in most cases, a death trap, as, with plastic walls that are impossible to climb, they rarely have ramps to facilitate exit. This article describes the design of a proposed edge-computing module that, using embedded vision, identifies the fall of people and animals in irrigation reservoirs. The module includes 180-degree panoramic cameras with colour night vision capability and an NVIDIA Jetson Orin Nano Super. The lack of databases covering the problem to be solved has been addressed by generating synthetic videos showing animals or people falling into irrigation reservoirs. The effectiveness of the training carried out using these synthetic sequences has subsequently been successfully validated using images captured in real-world environments. Full article

Environmental monitoring in hazardous indoor zones conventionally relies on fixed-sensor networks or manual inspections, both of which suffer from spatial blind spots and increased human exposure risks. This paper addresses the problem of transforming sparse, mobile sensor measurements into spatially resolved risk assessments in GPS-denied environments. We propose a Hazard Index (HI) framework that normalizes environmental parameters against established safety thresholds into a unified, graduated risk metric with $O\left(N\right)$ computational complexity, where N is the number of monitored parameters. The framework is designed for multi-parameter hazard assessment; the present work validates the computational pipeline, spatial mapping methodology, and classification logic through single-parameter CO₂ detection ($N=1$) deployed on a LiDAR-guided robotic platform integrating an MQ-135 gas sensor interfaced via a NodeMCU ESP8266 microcontroller. Experimental validation across a 144 sq ft indoor area achieved a trajectory-following RMSE of 0.54 ft relative to planned waypoints using Hector SLAM without odometry, detected CO₂ concentrations ranging from 0.02% to 0.25%, and identified a hazardous region encompassing eight measurement points ($\mathrm{HI}\ge 1.0$) using a three-tier classification scheme (Safe, Elevated, Hazardous) within 225 s of active mapping. The framework provides a lightweight computational footprint suitable for real-time evaluation on an NVIDIA Jetson Nano. The proposed approach establishes a cost-effective, reproducible methodology for autonomous indoor environmental monitoring, with the modular architecture designed for future expansion to multi-parameter sensing. Full article

Accurate wheat spike detection is essential for crop phenotyping and yield estimation, but real-world field conditions—such as dense spike overlap, environmental domain shifts, and degradation-induced failures like motion blur—pose significant challenges. Achieving robust perception under these circumstances while maintaining a strict accuracy-efficiency trade-off for edge devices remains a pressing research problem. To overcome these limitations, we propose GG-YOLO, a unified lightweight detection framework specifically tailored for complex agricultural environments. Rather than a simple recombination of existing lightweight modules, GG-YOLO integrates three original structural adaptations: First, a Dual-path Attentive Ghost Mechanism (DAGM) introduces gradient-guided attention modulation to enhance feature discrimination and explicitly resolve feature confusion in dense, overlapping regions. Second, a C3Ghost module combines multi-branch aggregation with linear feature generation, mitigating parameter redundancy in the prediction head by approximately 31% compared to the standard YOLOv8s without sacrificing semantic capacity. Third, DSample, a dynamic upsampling operator featuring an original dual-mode adaptive mechanism, robustly recovers fine-grained spatial details during multi-scale feature pyramid fusion. Extensive cross-dataset experiments on the GlobalWheat2020 and HNKJXYwheat datasets validate the model’s exceptional resilience to domain shifts and varying growth stages. GG-YOLO achieves a precision of 94.35%, a recall of 91.93%, and a state-of-the-art mAP@50 of 96.47%. Furthermore, the model contains only 7.89 M parameters and requires 20.4 GFLOPs, reaching an inference speed of 165 FPS on a desktop GPU and a validated real-time speed of 64 FPS on an NVIDIA Jetson edge computing platform. These results demonstrate that GG-YOLO establishes a superior accuracy-efficiency frontier, making it highly reliable for real-time field deployment in precision agriculture. Full article

Aiming at the core pain point that in existing object detection models, it is difficult to balance detection accuracy and real-time inference efficiency on edge computing devices in UAV-based intelligent inspection of power transmission lines, this paper proposes a lightweight YOLO-PowerLiteV2 model for anomaly target detection in power transmission lines to address the shortcomings of YOLO-PowerLite. Based on YOLO11n as the baseline, the model achieves compression of model volume while guaranteeing detection performance through four core improvements: the C3k2-UIB lightweight backbone module, the MCA (Multi-scale Cross-Axis) attention mechanism, the MBConv lightweight detection head, and the MFM (Modulation Feature Fusion) module. Experiments were conducted on a dataset constructed from 5563 aerial images of transmission lines containing three types of targets: bird nests, defective insulators, and balloons. The results show that YOLO-PowerLiteV2 achieves a mAP@50 of 95.2%, with only 0.97 M parameters and 2.8 G floating point operations (FLOPs). Compared with the baseline model, the number of parameters is reduced by 62.5%, and FLOPs are decreased by 56.25%. On the NVIDIA Jetson Xavier NX edge platform, the model achieves 59.5 FPS with only 16.8 ms latency, outperforming the baseline by 31% in frame rate. Its comprehensive performance outperforms mainstream lightweight detection models. The model demonstrates excellent adaptability to UAV edge-terminal deployment requirements, thereby providing technical support for real-time intelligent inspection of power transmission lines. Full article

Real-time video object detection on unmanned aerial vehicles (UAVs) is essential for urban inspection and autonomous perception, yet its deployment on edge devices is severely constrained by the high computational cost of accurate detectors, the quantization sensitivity of hybrid convolution-attention networks, and the system-level latency of full video processing pipelines. To address these challenges, we present DUST-YOLO, a deployment-oriented algorithm-hardware co-design framework, where structured pruning and mixed-precision quantization-aware training (QAT) are jointly optimized with TensorRT–DeepStream for efficient UAV small-object detection on edge platforms. First, we introduce a multi-dimensional structured pruning strategy that applies asymmetric channel pruning to convolutional and feature-fusion modules while compressing the Swin Transformer prediction heads and bottleneck stacks, thereby reducing parameters and computation with limited impact on multi-scale representation capability. Second, we develop a hardware-aware mixed-precision QAT scheme that maps computation-intensive backbone layers to INT8 while preserving the Transformer-related modules in FP16, improving inference efficiency while mitigating the accuracy loss caused by uniform low-bit quantization. Third, we compile the optimized network with TensorRT and integrate the resulting inference engine into a DeepStream-based asynchronous video pipeline on the edge platform, enabling end-to-end acceleration by reducing decoding, preprocessing, and memory-transfer overheads. Experimental results on the VisDrone2019-DET dataset and the NVIDIA Jetson Orin NX demonstrate that DUST-YOLO achieves 43.7% mAP@0.5 accuracy with an end-to-end latency of 36.3 ms and a throughput of 27.5 FPS. Compared with the state of the art, DUST-YOLO reduces end-to-end latency by 56.9% and improves end-to-end video throughput by $2.31\times$. Full article

Local feature matching plays a critical role in robotic SLAM and visual localization. However, in weakly textured indoor industrial environments, lightweight appearance-based methods often struggle to learn discriminative and stable local features. To address this challenge, this paper proposes GAEFeat, short for Geometry-Aware Efficient Feature, a lightweight vision–geometric feature learning network. To address the scarcity of specialized training data, we integrated robotic arm pose priors with depth information to automatically generate cross-view supervision signals and surface-normal labels. Based on this strategy, we constructed two complementary datasets, including a simulated dataset and a real-world dataset, to support feature learning and evaluation in weakly textured indoor industrial environments. For feature extraction, we design a dual enhancement mechanism consisting of a geometric auxiliary branch and a geometry-aware enhancement (GAE) module. The former guides the network to perceive local surface structures through surface normal supervision, while the latter utilizes a gating mechanism to achieve deep fusion between geometric priors and 2D texture descriptors. Experimental results demonstrate that GAEFeat achieves strong robustness and high inference efficiency in relative pose estimation, homography estimation, and visual localization tasks, with particularly notable advantages in near-field, weakly textured industrial scenes. The framework achieves an inference latency of only 3.9 ms on the NVIDIA Jetson AGX Orin edge platform, demonstrating its real-time capability and practical potential for deployment in edge computing environments. Full article

Global Navigation Satellite Systems (GNSS)-denied environments pose significant challenges for autonomous drone navigation, requiring robust visual localization systems capable of real-time performance. Existing approaches either sacrifice accuracy for speed or fail to adapt to varying flight altitudes and orientations, limiting their practical deployment. We present Real-Time Elevation and Orientation-Aware Localization Architecture (REOLA), a visual localization system that combines similarity-driven autonomous window sizing, element-wise correlation-based orientation detection, and reinforcement learning with human feedback (RLHF) enhancement for publicly available satellite imagery. On desktop hardware (i7-10700K + RTX 3070), the REOLA achieved approximately 59 FPS performance with sub-5-m accuracy across diverse flight conditions through intelligent similarity-based matching, combined with efficient MobileNet-V3 embeddings and FAISS similarity search. For embedded deployment on NVIDIA Jetson Orin Nano, the system achieved 22.5 FPS, meeting real-time requirements for autonomous drone localization. The system autonomously selects optimal window sizes corresponding to the current elevation and determines drone orientation through element-wise correlation scoring across discrete rotation angles. Enhanced through RLHF, the REOLA achieved a $97.1\%$ success rate (sub-5-m localization) while processing frames in 17 milliseconds on desktop hardware (44.4 ms on embedded hardware), providing a substantial margin over real-time requirements. The approach demonstrates particular superiority over traditional keypoint-based methods in challenging environments with repetitive patterns such as agricultural fields, rocky mountains, dense forests, and grasslands, where conventional keypoint detection struggles. We explicitly identify featureless sand dune deserts and open-sea or coastal water flights as out of scope, since the reference satellite imagery in those regimes does not contain stable landmarks. Full article

Anti-UAV surveillance demands real-time pixel-level UAV localization on resource-constrained gimbal-mounted platforms, yet existing lightweight segmentation models suffer from low recall that propagates to downstream tracking failure. Building on our prior dataset of 605,045 paired visible-light and infrared images, we extend the lightweight ThinDyUNet baseline with three architectural improvements: (1) symmetric dynamic convolution applied to both the encoder and decoder, (2) attention gates filtering skip connections, and (3) deep supervision with auxiliary loss heads. The resulting DAS-Net is evaluated under a three-seed Monte Carlo cross-validation protocol on the full 174,008-image test set. DAS-Net achieves a mean test mIoU of 0.6780 and Dice coefficient of 0.7509 across three independent seeds, outperforming the ThinDyUNet baseline by +6.65 percentage points (pp) in mIoU with statistical significance (one-sided paired t-test, p = 0.045, Cohen’s d = 1.74; full variance and significance analysis in the experimental section). DAS-Net matches the best-performing external baseline (UNet) and exceeds the others (MobileUNet, PAN, PSPNet) while using approximately 14.7× fewer parameters than ResNet-34-based variants. DAS-Net runs at 8.83 ms per image on an NVIDIA A6000 GPU (113 FPS) and 38.44 ms on an NVIDIA Jetson AGX Orin (26 FPS at FP16), demonstrating real-time deployability across server-class and embedded edge platforms. Full article

This paper presents a thermal and power characterization of the NVIDIA Jetson AGX Orin deployed in a realistic outdoor edge AI enclosure, integrated with power supplies, a Long Term Evolution (LTE) module, and a thermostat-controlled fan, across an ambient temperature range from −20 °C to +40 °C. The device was tested in a climate chamber under two workloads: a synthetic CPU and GPU stress test, and a YOLOv8s inference workload (TensorRT FP16, 640 × 640 input). Internal temperatures were recorded using four calibrated platinum Resistance Temperature Detectors (RTD)—PT100/PT1000, while Jetson chip temperatures and power consumption were logged via tegrastats and jtop. At sub-zero ambient temperatures, heat dissipated by the device itself kept all components within their operating ranges down to −20 °C. At the +40 °C setpoint, the stress test triggered Jetson thermal throttling at GPU and CPU temperatures of +95.6 °C and +99.0 °C, respectively. Under the same conditions, the YOLOv8s inference workload sustained 108.8 frames per second (FPS) at 19.1 W average power, approximately half of the 36.5 W consumed under the stress test, with chip temperatures well below the throttling threshold. These findings indicate that synthetic stress tests substantially overestimate the thermal and power demands of the tested inference workload, and that the enclosure retains sufficient thermal headroom for outdoor edge device deployment. Full article

Large language models (LLMs) are increasingly considered for deployment on edge and limited-resource systems, where local inference can reduce latency, improve privacy, and decrease dependence on cloud infrastructure. While prior studies have evaluated either task accuracy or hardware efficiency in isolation, few benchmarks combine generation-based response-quality evaluation with real-device power measurements on a representative limited-resource platform. This study addresses that gap by benchmarking twelve compact and mid-scale open-weight LLMs (sub-1B to 8B parameters), evaluating generation-based accuracy on a desktop platform and measuring deployment efficiency—throughput, power consumption, and energy use—on an NVIDIA Jetson Orin Nano Super; the accuracy–efficiency trade-off is thus established at the model-configuration level. Unlike prior Jetson-based evaluations relying solely on internal telemetry, this work pairs generation-compatible lm-eval accuracy tasks with a dual power-measurement setup that combines internal tegrastats rail readings with external board-level input power measured using a digital multimeter and explicitly compares GPU-accelerated and CPU-only inference modes. GPU-accelerated inference provided a clear advantage, increasing median throughput from 7.12 to 18.13 tok/s and improving external-meter energy efficiency from 0.453 to 0.823 tok/J, despite higher mean input power. Sub-1B models offered the best throughput and energy efficiency, whereas 7–8B models achieved stronger accuracy at a substantially higher energy cost per generated token. These results demonstrate that edge LLM deployment requires multi-objective evaluation balancing accuracy, throughput, power consumption, and energy efficiency. Full article

The increasing penetration of renewable energy sources in inverter-dominated microgrids introduces significant challenges for maintaining voltage and frequency stability under weak-grid and dynamically varying operating conditions. Conventional inverter control strategies, including droop control and virtual synchronous machine (VSM) methods, often exhibit limited adaptability and degraded transient performance under renewable intermittency and uncertain load variations. This paper proposes a physics-informed neural-network (PINN)-based supervisory framework for real-time grid-forming inverter control. The proposed approach embeds swing-equation dynamics, Kirchhoff-based electrical constraints, and stability-aware objectives directly into the neural-network optimization process to improve physical consistency, robustness, and operational reliability. The controller is trained offline and deployed for low-latency online inference on an NVIDIA Jetson AGX Xavier embedded platform. Simulation and hardware-in-the-loop validation results demonstrate improved transient stability, reduced frequency deviation, enhanced voltage regulation, and superior robustness compared with conventional droop, VSM, and purely data-driven neural-network controllers. The proposed framework achieved an average inference latency of approximately 0.7 ms while maintaining stable operation under renewable intermittency, load disturbances, and weak-grid conditions. The results demonstrate the potential of physics-informed machine learning for supervisory real-time control of inverter-dominated microgrids and intelligent renewable energy systems. Full article

This paper addresses the problem of efficient federated learning in energy-harvesting AIoT systems, where time-varying energy availability may lead to device blackouts and unstable learning performance. To address this issue, we propose an energy-adaptive multi-dimensional learning control framework that jointly determines model complexity and training intensity based on the real-time energy state of each device. This method integrates multiple control dimensions, including model pruning, quantization, knowledge distillation, and adaptive local training, into a unified decision mechanism under an energy constraint. Each device determines its participation in federated learning based on its residual energy relative to an energy threshold. When participating, the device selects a feasible learning configuration that jointly considers training intensity (e.g., epoch size and batch size) and lightweight learning operations to maximize learning effectiveness while preventing energy depletion. The proposed framework was implemented on a real-world testbed using NVIDIA Jetson Orin Nano devices under solar-energy-harvesting conditions. Our experimental results demonstrate that the proposed method significantly reduces device blackout while maintaining competitive model accuracy with respect to energy-unconstrained scenarios. These results highlight that joint control of multiple learning-cost factors is essential for achieving stable and efficient federated learning in energy-harvesting AIoT environments. Full article