Search Results (2,878)

Hyperspectral anomaly detection (HAD) identifies targets by spectral differences. However, large deep learning models overfit under the small-sample conditions typical of remote sensing, where anomalies are sparse and annotations costly. We propose a lightweight network on the GT-HAD transformer backbone for hyperspectral imagery. The design includes: (1) capacity reduction via multi-layer perceptron (MLP) ratio and embedding dimension optimization; (2) a Decoupled Projection Gating Fusion (DPGF) module enforcing spatial–spectral decoupling via dual-path projection and complementary regularization; (3) analysis of capacity–generalization tradeoffs across six airborne hyperspectral datasets. Experiments reveal an inverted-U relationship between capacity and generalization. The Mini configuration (121 K parameters) achieves a 53% parameter reduction. It yields AUC improvements on three datasets, maintains negligible performance loss (<0.1%) on two datasets, and exhibits a measurable decline on only one dataset, compared with the 255 K baseline. Notably, the baseline attains higher accuracy with only 25% training data on Pavia (+3.69% area under the receiver operating characteristic curve, AUC). This dataset-specific observation provides empirical evidence that capacity–data mismatch can induce severe overfitting in deep HAD models. Under a 25% training ratio, DPGF shows observable performance trends on spectrally homogeneous scenes, and zero initialization ensures identical performance to that of the baseline at the initial training stage, eliminating insertion risk during module deployment. However, the limited diversity of sensors and scene types constrains the generalizability of the observed trends. These results demonstrate that spatial–spectral decoupling with lightweight design suppresses overfitting in few-shot HAD, guiding compact models for onboard satellite and unmanned aerial vehicle (UAV) applications. Full article

Despite the rapid proliferation of unmanned aerial vehicles (UAVs) across industrial, agricultural, and scientific domains, their deployment remains constrained by limited endurance, aerodynamic inefficiency, and acoustic emissions, all mediated by a shared onboard energy budget. Existing biomimetic UAV reviews have generally treated aerodynamics, structures, sensing, control, and energy systems as parallel topics rather than as interacting components of a unified aerial architecture. Drawing primarily on literature published between 2015 and June 2026 and identified through searches of Web of Science, Scopus, and Google Scholar, this review addresses this gap by examining bio-inspired technologies across six principal domains: aeroacoustic and passive flow control, aerodynamic efficiency, multifunctional structural composites, neuromorphic sensing and control, ionic energy storage, and energy harvesting. Its principal contribution is a cross-domain synergy analysis identifying five performance couplings and one structural enabling architecture through which these domains interact physically and functionally. Representative examples include serration-based propeller geometries that can simultaneously reduce noise and power demand; morphing wing surfaces that serve as both aerodynamic structures and triboelectric harvesting substrates; and neuromorphic spiking neural networks that have been reported, in specific event-vision inference benchmarks, to reduce inference energy by three to four orders of magnitude relative to embedded graphics processing unit (GPU)-based implementations. Mechanical harvesting outputs nonetheless remain orders of magnitude below propulsion requirements and are thus positioned as supplementary. Four systemic barriers (unquantified mass–energy balance, undocumented durability, aeroelastic co-design gaps, and heterogeneous metrics) are evaluated, and the resulting synthesis indicates that advancing bio-inspired UAVs requires a transition from structural imitation to functional, system-level biomimetics. Full article

Driven by the vision of sixth-generation (6G) communication networks, Space–Air–Ground Integrated Vehicular Networks (SAGVNs) address the connectivity blind spots inherent in traditional networks by integrating unmanned aerial vehicles (UAVs) as highly mobile relay nodes. However, the high bit error rates (BERs) and prolonged propagation delays characteristic of satellite links, coupled with the highly dynamic topologies and multi-hop transmission nature of UAVs and terrestrial vehicles, present significant challenges to reliable end-to-end data streaming. To mitigate the performance degradation caused by link asymmetries in heterogeneous networks, this paper proposes a reliable transmission optimization scheme for UAV-relayed SAGVNs. By comprehensively modeling the transmission dynamics of long-delay, high-BER satellite links and mobile multi-hop UAV networks, the proposed scheme introduces an enhanced slow-start mechanism to accelerate throughput growth, thereby mitigating the startup lag induced by extensive propagation delays. Furthermore, an accurate packet loss differentiation model is established during the congestion avoidance phase. This model effectively decouples non-congestion packet losses—triggered by random channel errors or topology handovers due to high-speed node mobility—from genuine congestion-induced losses caused by buffer overflows at bottleneck nodes. Simulation results demonstrate that the proposed adaptive scheme demonstrates notable improvements over classical loss-based and delay-based baselines in reducing queuing delays at UAV relay nodes, enhances the transmission efficiency of multi-hop terminals, and effectively maintains end-to-end goodput stability in high-latency environments. Full article

Cross-view geo-localization (CVGL) is a critical technology used in unmanned aerial vehicles (UAVs) and widely applied in navigation and target localization tasks. However, owing to the extreme perspective disparity between UAV oblique views and satellite vertical views, CVGL still involves significant challenges, including geometric distortion caused by viewpoint differences, drastic appearance inconsistencies, and the difficulty in bridging semantic gaps between heterogeneous data. To address these issues, we propose a novel CVGL method named dynamic-feature collaborative optimization and semantic-alignment network (DFSA), designed to extract robust feature representations and achieve fine-grained alignment. Specifically, the DFSA employs a residual-based vision transformer as the backbone to capture global context while alleviating the training instability and feature collapse often associated with standard transformers. To bridge the semantic gap between global and local features, we design a feature optimization module comprising a local feature enhancer and a global feature aggregator. This module establishes a closed-loop collaborative system that facilitates top-down semantic guidance and bottom-up detail feedback. Furthermore, we introduce a semantic segmentation and alignment module that adaptively partitions images into semantic regions based on feature response distributions, shifting the matching granularity from the global level to the semantic region level to effectively overcome feature mismatches caused by positional offsets and scale variations. Extensive experiments conducted on the University-1652 and SUES-200 datasets demonstrate the superior image retrieval performance of the proposed DFSA. Specifically, DFSA achieves a Recall@1 of 94.87% and an Average Precision (AP) of 95.32% on the University-1652 dataset and maintains highly competitive Recall@1 performances between 96.83% and 99.25% across various altitudes on the SUES-200 dataset. These results validate the model’s effectiveness in handling extreme viewpoint changes for UAV-based cross-view image retrieval tasks. Full article

High-precision drogue localization during terminal guidance is critical to close-range autonomous unmanned aerial vehicle (UAV) docking and hinges on infrared–depth (IR–D) multimodal detection. Yet, deploying such detection on airborne edge computing platforms faces severe challenges due to modal heterogeneity, feature redundancy, and real-time constraints. A lightweight IR–D fusion detection network, termed AWIE-CGAF, is proposed for airborne edge deployment, which integrates frequency-domain, physics-prior-driven input enhancement with decoupled gated attention-based adaptive feature fusion to achieve efficient multimodal detection. A training-free Adaptive Wavelet Image Enhancement (AWIE) module is designed to differentially modulate image structures and details in the frequency domain, improving the signal-to-noise ratio and feature discriminability. Concurrently, a Cross-Gated Attention Fusion (CGAF) module employs decoupled cross-modal attention with independent gating, preserving modality-specific features while dynamically selecting complementary information, mitigating redundancy and feature contamination. Experiments on the self-constructed Drogue Infrared–Depth (DIRD) dataset showed that AWIE-CGAF achieved 89.5% mAP@0.5 and 58.2% mAP@0.5:0.95 with 13.5 M parameters, while maintaining real-time inference at 51.7 FPS on a Jetson AGX Orin edge platform. Among the evaluated methods, the proposed framework achieved the highest detection accuracy while retaining real-time edge inference capability. These results support the feasibility of AWIE-CGAF for resource-constrained IR–D drogue perception. Full article

Wireless sensing networks increasingly extend into obstacle-prone deployments, where physical blockage degrades reliability and open propagation exposes transmission activity. Intelligent reflecting surfaces (IRSs) establish programmable paths around obstacles while passive elements remain constrained by severe cascaded attenuation. To address the tradeoff between reliability and covertness, we propose an unmanned aerial vehicle (UAV) -assisted active-IRS architecture under probabilistic line-of-sight and non-line-of-sight propagation conditions that accounts for direct leakage from the transmitter to the warden together with residual jammer cancellation and always-on IRS circuit noise under a finite output power budget. Furthermore, bidirectional Kullback–Leibler analysis identifies the reverse divergence as the tighter restriction and converts the covertness requirement into conservative gain bounds under warden location uncertainty and relative phase uncertainty conditions between the direct and aggregate reflected fields. Subsequently, closed-form phase control for calibrated equal-gain elements and gain monotonicity reduce the joint design to an exhaustive search over the prescribed placement grid. The numerical results demonstrate a SINR advantage over passive reflection and single-element relaying across the evaluated settings. The finite-array and hardware analyses show that gain back-off enforces a prescribed covert-outage limit while direct leakage and residual self-interference remain explicitly controlled. Overall, the framework provides a transparent basis for reliable covert sensing through UAV-assisted active reflection. Full article

Optimal sensor placement (OSP) is critical to building a full-lifecycle health monitoring network for high-aspect-ratio unmanned aerial vehicle wings, as it directly affects the deformation sensing of the wing shape. To address this challenge, this paper proposes a novel sensor placement method for wings based on a chaotic strengthened aquila optimizer (CSAO) that integrates chaotic mapping and a nonlinear search strategy. Specifically, the proposed method introduces a uniform initialization strategy based on the piecewise chaotic map and a nonlinear criterion for switching between exploration and exploitation in the basic aquila optimizer (AO). These enhancements increase the diversity of the initial population and raise the probability of global search in later iterations, thereby accelerating convergence and strengthening global optimization capability. First, the performance of the CSAO is compared with that of other popular intelligent algorithms on 10 benchmark functions. The results show that the proposed method exhibits superior convergence speed, higher-quality solutions, stronger global search ability, and better robustness, making it suitable for OSP problems involving tens of thousands of candidate points. Next, the CSAO is applied to sensor placement on a wing-shaped plate. Compared with other OSP methods, the proposed method offers significant advantages in terms of sensor distribution, computational time, and hardware cost. Finally, experimental validation is conducted using a wing test platform equipped with fiber Bragg grating (FBG) strain sensors. The measurement results demonstrate that the reconstructed shape is in excellent agreement with the measured shape. Therefore, the proposed CSAO-based OSP method, combined with the FBG-based structural monitoring system, offers a promising solution for health monitoring of deformable structures in extreme environments. Full article

Fingertrap is a network encryption and authentication protocol that extends the X3DH and Double Ratchet frameworks with three novel mechanisms inspired by the Chinese finger trap (zhĭ wăng): a friction ratchet that exponentially increases computational cost for each failed authentication attempt; a recursive annihilation protocol that irreversibly destroys all cryptographic state after a configurable failure threshold; and a commit-then-challenge handshake that requires a counterintuitive “inward” action for legitimate authentication. A bidirectional weave hash extends the Double Ratchet’s transcript binding to cover every message in both directions. Together, these mechanisms provide per-message forward secrecy, post-compromise security (self-healing), clock-free operation, and a self-destruct capability. The individual ingredients-client puzzles, key erasure, and ratcheting-each build on established lines of work; their combination into a single stateful protocol, in which failed authentication attempts cryptographically tighten the session state and ultimately destroy it, is not to our knowledge offered by deployed transport protocols such as TLS 1.3, Signal, or WireGuard. The design targets deployments in which interception or capture of a device implies endpoint compromise, such as Unmanned Aerial Vehicle (UAV) telemetry links and body-worn sensors, where denial of exploitation requires guaranteed loss of past and future session material. We describe the full protocol, provide game-based security arguments under an explicit adversarial model, give analytic cost estimates for the friction mechanism, analyse the denial-of-service surface and a two-layer mitigation strategy, and specify a post-quantum extension using hybrid $\mathsf{X25519}\text{/}\mathsf{ML}\text{-}\mathsf{KEM}\text{-}\mathsf{768}$ ratcheting. Full article

Flying Ad Hoc Networks let unmanned aerial vehicles communicate directly, without relying on fixed ground infrastructure, in time-critical settings such as disaster response, search and rescue, border surveillance, precision agriculture, and military reconnaissance. UAV mobility, however, causes frequent topology changes and intermittent connectivity, so nodes typically store data locally and replicate it across the network to keep it available. The resulting challenge is consistency: independently evolving copies must be reconciled without a central coordinator, and strong consistency is not realistic in a network this prone to partitioning. We address this with the Decentralized Mobile Reconciliation Protocol (DMRP), which provides eventual consistency among UAV nodes with no external coordination, with convergence formally guaranteed whenever the swarm’s synchronisation graph is eventually connected. DMRP combines immediate local validation and convergence guarantees grounded in conflict-free replicated data type properties; hysteresis-based memory management with dual thresholds to cap journal storage overhead; adaptive delta or full-state synchronisation based on receiver lag; and epidemic propagation for transitive update dissemination. Energy efficiency guided the design throughout, through wireless broadcast and the avoidance of redundant transmissions. DMRP was implemented and evaluated through extensive OMNeT++/INET simulations of three-dimensional FANET scenarios. Results demonstrate that the protocol maintains a strictly bounded reconciliation journal, whereas the reference $\delta$-CRDT log grows without bound, reducing reconciliation-journal storage by up to 75% at the largest workload evaluated, while achieving near-complete consistency after node isolation and network partitioning. Full article

Unmanned aerial vehicle (UAV) accidents pose growing challenges to public safety and airspace management. This study develops an open-data-driven hierarchical Bayesian network (BN) with a Leaky Noisy-OR mechanism to analyse factors associated with consequence severity among recorded UAV accidents, and extends it to a Bayesian decision network for response simulation. Using 633 public accident records and matched meteorological data, 14 binary risk-factor nodes spanning human, machine, environmental, and management dimensions were constructed. Stratified five-fold cross-validation yielded a mean validation F1 score of 0.922 and an AUC of 0.784. Backward inference ranked airspace exposure, wind, and operation error highest under severe-consequence conditioning, whereas sensitivity analysis identified wind, bad weather history, and operation error as the most influential root-node parameters. Under the assumed directed acyclic graph (DAG), the bad weather history→weather→environment→risk state path had the highest average edge-influence score (0.853). Under the baseline safety-priority assumptions, the reroute strategy was preferred, yielding the highest expected utility (31.967) and reducing the model-estimated post-decision high-risk probability from 81% to 45%. Alternative preference settings ranked the adjust strategy first. The framework integrates open-data severity analysis with assumption-explicit response simulation. Full article

Missing bolts compromise the structural integrity of bolted connections in steel bridges and industrial infrastructure. Manual visual inspection remains labour-intensive, subjective, and hazardous in hard-to-reach locations. This study presents a comparative benchmarking framework for unmanned aerial vehicles (UAVs) missing-bolt patch classification using convolutional neural network (CNN), focusing on balancing accuracy and computational efficiency. A UAV-acquired dataset of bolt-centric image patches was developed to evaluate four systematic experimental schemes: (i) a custom lightweight CNN trained from scratch, (ii) the lightweight CNN integrated with Squeeze-and-Excitation (SE) attention blocks across multiple positions, (iii) nine fine-tuned state-of-the-art (SOTA) pretrained CNN backbones, and (iv) SE-enhanced versions of these pretrained models. All architectures were evaluated under a standardised experimental protocol. Results show that the proposed lightweight CNN achieves classification performance comparable to heavyweight pretrained models while requiring significantly lower computational resources. Integrating SE blocks did not improve classification performance for this localised task and, in several configurations, reduced accuracy and training stability. Pretrained transfer learning models achieved high accuracy overall, but their computational complexity limits direct deployment on edge devices and UAV platforms. Grad-CAM visual explanations confirmed that the lightweight CNN consistently focuses on relevant bolt and hole regions. The findings demonstrate that a task-specific lightweight CNN offers a practical balance between inspection reliability and deployment efficiency for automated structural monitoring. Full article

To address the challenge of rapid and precise obstacle avoidance for unmanned aerial vehicles (UAVs) in complex urban environments, rugged canyons, and other unstructured environments, this paper proposes a vision-based navigation algorithm. By combining the strengths of deep reinforcement learning (DRL) and convolutional neural networks (CNNs), this algorithm enables efficient navigation and obstacle avoidance in dynamic environments. First, to improve training efficiency, an autoencoder is used to extract latent spatial vectors from depth images, which are then used as input features for DRL. Second, an artificial potential field (APF) is introduced into the reward function to enhance obstacle avoidance performance in dynamic environments. Third, a CNN-based adaptive mode-switching mechanism is designed to meet navigation requirements under different environmental conditions. This mechanism can automatically identify environmental features based on real-time input data and dynamically adjust the UAV’s navigation strategy. To evaluate the proposed method, simulation experiments were conducted in static and dynamic scenarios, together with a preliminary indoor flight test. Under the evaluated conditions, the proposed method achieved favorable navigation success rates and path efficiency compared with the selected visual DRL baselines. The results also indicate cross-scenario transferability to the tested environments without environment-specific retraining. Full article

Multi-agent reinforcement learning has emerged as a promising paradigm for cooperative unmanned aerial vehicle (UAV) swarm coordination. However, existing communication-aware MARL methods primarily focus on communication topology, message routing, and message aggregation, while the information content of the exchanged messages is often only implicitly controlled. In realistic UAV networks, inter-agent communication is constrained by limited bandwidth, communication range, energy consumption, and packet loss. It is therefore desirable for each UAV to transmit compact and task-relevant information rather than dense and redundant latent features. In this paper, we propose IB-CEMARL, an information-bottleneck-guided, communication-efficient multi-agent reinforcement learning framework for UAV swarms. We formulate inter-UAV communication as a minimal sufficient message-learning problem in which each UAV encodes its local observation into a stochastic bottleneck message before exchanging information with its neighbors. Cauchy–Schwarz divergence-based quadratic mutual information is adopted as a unified dependence measure to jointly regularize message compression, preserve decision-relevant information, and reduce statistical redundancy among neighboring UAV messages. Extensive experiments demonstrate that IB-CEMARL achieves superior cooperative performance, reduced message redundancy, and stronger robustness compared with representative communication-aware MARL baselines. In particular, IB-CEMARL improves the average return by $4.9\%$ and reduces inter-message dependence by $29.0\%$ compared with the KL-IB-MARL baseline while maintaining efficient communication under constrained bandwidth settings. Full article

Existing reviews on AI in tea production are either agriculture-generic or limited to isolated tasks. This review thoroughly compares vision technologies (RGB, hyperspectral, near-infrared, thermal, Light Detection and Ranging (LiDAR), Unmanned Aerial Vehicle (UAV)) and establishes a task-oriented algorithm selection framework for the tea industry. For small-sample or near-linear problems, traditional machine learning (ML) (support vector machine (SVM); partial least squares regression (PLSR)) remains effective. For unstructured field tasks, deep learning achieves superior performance: pest detection accuracy exceeds 97%, tea bud detection reaches 96.8% with RGB images, and hyperspectral imaging predicts nitrogen content with R² > 0.90 and tea polyphenols with R² up to 0.925. Algorithm choice further differentiates by task granularity: lightweight convolutional neural networks (CNNs) balance speed and accuracy for edge deployment at 16 fps; You Only Look Once (YOLO) series detectors enable real-time localization on mobile platforms at 93.1% accuracy, 24 ms per target. No single algorithm dominates all tea tasks; selection is a trade-off among accuracy, speed, data availability, and computational constraints. These findings outline a structured analysis of the challenges and pathways for transitioning computer vision (CV) from laboratory research toward field-deployable tools. Full article

To address the scarcity of bridge defect data in the harsh environments of cold regions, as well as the parameter redundancy and edge platform deployment challenges of existing algorithms, this paper proposes an intelligent inspection system for bridge surface defects in cold regions based on parameter sharing and feature enhancement. The system first constructs a large-scale dataset called CRBD (Cold-Region Bridge Defect), which contains 10,129 high-resolution images and finely classifies defects into four standardized categories: Crack, Spalling, Patch, and Seepage. Subsequently, a lightweight detection network called BridgeNet is designed. Its core parameter sharing and feature enhancement detection head stabilizes training via group normalization, significantly reduces the parameter count through cross-scale global sharing and structural reparameterization, and improves bounding-box regression accuracy by incorporating a distribution focal loss mechanism. On this basis, an airborne real-time image processing and intelligent perception pipeline is constructed, which establishes the complete workflow for autonomous unmanned aerial vehicle inspections. The experimental results demonstrate that with a lightweight architecture of only 2.26 M parameters and a model size of 4.98 M, BridgeNet achieves a mean Average Precision of 61.4% and an F1 Score of 60.9%. Furthermore, it exhibits excellent real-time inference speed on heterogeneous edge mobile platforms and maintains robust overall perception stability under various extreme physical disturbances. Full article