Search Results (308)

UAVs increasingly rely on accurate SLAM for aerial mapping and inspection in GPS-denied environments. 3D Gaussian Splatting (3DGS) has opened a new direction for UAV mapping by allowing SLAM systems to build dense, photorealistic, and renderable maps. Yet in 3DGS SLAM the map is optimized from the pose graph, so a false loop closure can deform both the UAV trajectory and the Gaussian map consumed by downstream UAV autonomy. Reliable loop admission is therefore relevant to safe GPS-denied operation because it protects the state and map estimates on which autonomous functions depend. The present work evaluated this upstream estimation-integrity problem; it did not measure closed-loop guidance, control, or navigation-safety outcomes. We address the loop-admission problem that arises after a place-recognition (PR) module proposes a candidate loop and relative-pose seed. GSSeq is a rendered-reference sequential verifier that uses the current Gaussian map as active evidence before inserting a loop factor. It renders RGB-D references with the PR seed, checks LiDAR/rendered-depth consistency and image/rendered-reference consistency over active support, and propagates the seed through a short query trajectory window. A loop is admitted only when this evidence remains geometrically supported and photometrically stable. On fixed LiDAR-PR candidate sets spanning MARS-LVIG, MUN-FRL, and independent NTU-VIRAL aerial sequences together with ground-mobility benchmarks, GSSeq provides a competitive precision-oriented operating point while suppressing false loop admissions. Thresholds calibrated only on NTU-VIRAL spms_01 combine rendered RGB agreement with LiDAR-submap geometry and are then frozen for spms_02. On this held-out sequence, GSSeq rejects all seven false-positive BTC factors while retaining one of three true-positive factors. The trajectory-to-map experiment reduced ATE RMSE from $2.609\mathrm{m}$ to $1.417\mathrm{m}$ and improved selected-view PSNR from $13.80\mathrm{dB}$ to $16.46\mathrm{dB}$. These results show that rendered verification can preserve an aligned, renderable UAV trajectory-map pair before unsupported loop factors reshape the SLAM map. Full article

Virtual reality (VR) applications are often designed as episodic experiences, with limited support for persistence, longitudinal revisitation, and structured integration of interaction data across sessions. This paper presents a persistent multi-user VR garden architecture that combines snapshot-based state restoration, structured event, movement, and transcript records, an asymmetric owner–visitor workflow, cloud-mediated speech transcription, and deferred AI-supported synthesis. The architecture separates the current spatial configuration of the environment from the interaction traces through which it evolves, supporting repeated access, state restoration, historical consultation, and post-hoc processing. A controlled technical validation using synthetic or researcher-generated inputs was conducted through Unity Editor/backend tests and on Meta Quest 3 hardware. Persistence was evaluated at 20, 100, and 250 objects in the Editor and at 1, 50, and 100 objects on Quest, with two Quest replicas per load. Application-level visitor restrictions were verified across nine prohibited write operations. A frozen production speech-to-text corpus completed 40/40 requests with a micro-averaged word error rate of 9.50% and a median end-to-end latency of 1.619 s. The deferred AI pipeline was additionally verified as a functioning technical integration, while limitations in semantic-detail preservation were observed. The results support the implementation-level feasibility of the proposed persistent and traceable VR architecture. The present evaluation does not establish backend-level authorization guarantees, general AI or NPC-grounding performance, user outcomes, or clinical effectiveness. Full article

Small- and medium-sized landslides frequently occur in clusters and exhibit fragmented morphologies, irregular boundaries, and spectral characteristics similar to surrounding roads, bare soil, and sparsely vegetated surfaces, making their automated extraction from remote sensing imagery challenging. Although the Segment Anything Model (SAM) provides strong general-purpose segmentation capabilities, its direct application to landslide mapping is limited by the geoscience domain gap and its dependence on external prompts. This study proposes the Asymmetric Boundary-aware Segment Anything Model (AB-SAM), a parameter-efficient adaptation of SAM for automated landslide semantic segmentation. AB-SAM integrates three task-specific components. First, the offline Multi-Feature Variation-Guided Prompting (MF-VGP) module generates cached auxiliary bounding boxes from registered pre- and post-event images without accessing ground-truth masks. Second, the Asymmetric Feature Augmentation (AFA) strategy combines geometric perturbation, CutMix, and asymmetric dual-branch supervision, in which a Hint-free branch serves as the primary optimization pathway and a lower-weight box-guided branch provides auxiliary spatial supervision. Third, the Boundary-Aware Morphological Prompting (BAMP) module injects trainable boundary-aware morphological information into the largely frozen SAM image encoder. During validation, testing, and application, only the Hint-free branch is retained, enabling inference using post-event imagery without external point, box, or mask prompts. On the fixed, spatially disjoint Zixing test set, AB-SAM achieved an overall accuracy of 96.171%, a precision of 68.149%, a recall of 60.011%, an F1-score of 63.822%, a landslide-class Intersection over Union of 46.867%, and a mean Intersection over Union of 71.452%. Repeated experiments with three random seeds showed low run-to-run variation. Direct evaluation without retraining on the Hokkaido Iburi-Tobu dataset yielded a mean Intersection over Union of 66.136%, providing evidence of cross-region and cross-event transferability. These results demonstrate that AB-SAM provides a practical parameter-efficient framework for automated, hint-free landslide segmentation, although further evaluation across additional regions, sensors, and landslide-size distributions remains necessary. Full article

Permafrost maps are important for regional planning in cold regions, but binary products do not represent gradients in thermal conditions, seasonal thaw response, or mapping confidence. This study developed an uncertainty-aware regional permafrost-condition screening framework for the Daxing’anling region during 2003–2022 by integrating TTOP-derived mean annual ground temperature (MAGT), reconstructed permafrost-occurrence probability, and Kudryavtsev-model-derived active-layer thickness (ALT). MAGT ranged from −4.11 to 5.28 °C, with a mean of −0.28 °C, and comparison with 23 published borehole records from 17 reported locations yielded r = 0.756 and RMSE = 0.419 °C. Explicit measurement depths were available for 12 of the 23 records and ranged from 10 to 15 m. Model-derived ALT ranged from 1.480 to 1.864 m. The complete point-scale evaluation using all 12 valid maximum depth of seasonal thaw (MDST) records yielded an RMSE of 0.92 m. Adding ALT changed 21.05% of valid-pixel assignments. Cold–low-response permafrost, Cold–moderate-response permafrost, Warm–enhanced-response permafrost, Near-thaw transitional permafrost, Marginal/low-confidence permafrost, and Non-permafrost occupied 7.3%, 17.1%, 14.9%, 4.1%, 13.9%, and 42.8% of the domain, respectively. Full Monte Carlo uncertainty propagation retained 78.65% modal agreement with the deterministic baseline, and 56.65% of the domain had a maximum class-membership probability below 0.60, whereas threshold-only perturbation retained 98.92% agreement. The framework is therefore suited to regional investigation and monitoring prioritization; project-level engineering decisions require direct geotechnical and deformation-based evidence. Full article

Sea fog that lowers horizontal visibility below 1 km is a recurrent hazard to port operations and near-shore navigation, yet its objective, continuous short-range warning remains difficult. Using a microwave radiometer and a 35 GHz millimeter-wave cloud radar co-located at the Xiaoyangshan station near the Yangshan deep-water port, eastern China, supervised by a continuous minute-resolution visibility ground truth (about 0.65 million records, 2025–2026), we develop a calibrated probabilistic sea-fog nowcasting model for lead times of 0–3 h. Under strict date-grouped cross-validation—in which neither the input window nor any forecast label crosses a fold boundary—and at the operationally realistic fog base rate of ∼1.9%, the model attains a fog-state ROC-AUC of 0.95–0.96 across lead times and an onset-AUC of about 0.94 at the 3 h lead. After out-of-group isotonic calibration the output probabilities are reliable (expected calibration error 0.007), and a single-stage alarm reaches an event hit rate of 0.84 over the 2026 hold-out period—a figure that is unchanged under a strictly out-of-time protocol in which the model, the calibrator, and the alarm parameters are all frozen on data through 2025—with a median first alert about 3.1 h before fog onset. A compact near-surface scalar model already saturates discrimination; adding vertical profiles and radar microphysics through a mask-aware fusion network yields only a small, statistically non-significant gain, indicating that independent fog events, not model capacity, limit further improvement. The scheme is lightweight, locally deployable, and can be re-evaluated as observations accumulate. Full article

Seasonal freeze–thaw processes affect soil hydrothermal conditions in high-altitude valleys, yet evaluations based only on air temperature or maximum freezing depth may overlook the distinction between surface-connected freezing and delayed thawing within the soil profile. Meteorological conditions, ground surface temperature (GST), ground-temperature profiles, freezing depth, and volumetric water content (VWC) were continuously monitored in an arid valley on the Qinghai–Tibet Plateau. Mean annual GST was 3.23 °C higher than mean annual air temperature, and the freezing and thawing n-factors were 0.72 and 1.54, respectively, indicating weakened cold accumulation and enhanced heat accumulation at the ground surface. The maximum surface-connected freezing depth reached 3.30 m, whereas ground temperatures at 3.5 m and below remained above 0 °C. During spring thawing, a residual frozen layer persisted for 49 days after the shallow layer had thawed, with a maximum thickness of 3.24 m. GST-based freezing degree days represented freezing depth better than air-temperature-based freezing degree days. Soil VWC remained low, and precipitation responses were mainly confined to 0.2 m depth. These findings reveal a thermally dominated freeze–thaw regime with weak deep moisture response and show that distinguishing surface-connected freezing from residual frozen layers improves hydrothermal-state identification in low-water-content sandy gravel deposits. Full article

In order to improve the accuracy of near-ground temperature forecasting under winter rain, snow, and freezing weather conditions, three machine learning algorithms, namely neural network, random forest, and support vector machine, were used to train various ground and air elements in the rapidly updated assimilation model of Zhejiang Province based on multi-source observation data, reducing the error of temperature in the model field and forming hourly and 3 km horizontal resolution ground and air temperature datasets for two rainy, snowy, and frozen weather processes. After calibration using the backpropagation neural network algorithm, random forest algorithm, and support vector machine algorithm, the MAE of the simulated field temperature forecast decreased from 1.29 °C to 0.937 °C, 1.01 °C, and 0.988 °C, respectively. The backpropagation neural networks and support vector machine algorithms perform well, but support vector machine algorithms have relatively short computation times. Using 10 feature points for training achieves optimal performance; more points may not necessarily lead to better calibration results. Adding actual data at the initial time of the target point significantly improved the correction effect, and the improvement effect was even better when the forecast lead time was less than 10. The correction effect of the prediction field shows that when the forecast lead time is between 15 h and 24 h, it becomes unstable over time. The mean prediction accuracy of whether the temperature exceeds the 0 °C temperature threshold at 24 forecast moments before calibration is 0.928. After correction, it has been increased to 0.956. Full article

Ice lenses in natural frozen soils commonly exhibit inclined and heterogeneous distributions, and their spatial morphology significantly influences the mechanical behavior of frozen soils. To investigate the coupled regulatory mechanism of ice lens inclination angle and thickness on the mechanical properties of frozen silty clay, specimens containing artificial single-layer ice lenses with varying inclination angles (0°, 10°, 20°, 30°) and thicknesses (5 mm, 15 mm) were prepared under constant temperature, water content, and loading rate conditions. Low-temperature uniaxial compression tests were conducted, and the results were systematically analyzed in conjunction with discrete element method (DEM) simulations and a modified Duncan–Chang model. The results indicate that increasing the ice lens inclination angle leads to a nonlinear reduction in the deviatoric stress at 15% axial strain, with the failure mode transitioning from compression-induced bulging to shear sliding dominance. When the ice lens thickness increased from 5 mm to 15 mm, the deviatoric stress at 15% axial strain further decreased across all inclination angles, accompanied by a reduction in the composite modulus. The response surface prediction formulas for parameters a and b, established based on experimental data, effectively describe the stress–strain relationships. DEM simulations reveal, at the mesoscale, the asymmetric displacement field and shear band evolution mechanisms governed by inclined ice layers, with bond breakage accelerating as the inclination angle increases. This study clarifies the coupled effects of ice lens spatial configuration and confining pressure on the mechanical response of frozen soils, providing a theoretical reference for bearing capacity assessment of frozen ground containing inclined ice lenses. Full article

Quantifying how mirror-symmetric an image is about a given axis (symmetry scoring) underpins applications from visual aesthetics to medical imaging, yet proposed scoring methods have never been compared on a common, statistically grounded protocol. We benchmark 13 scoring methods (9 collected from the literature; 4 introduced here) spanning from classical features to frozen deep features, across four single-axis and five multi-axis datasets under a reflection-exact protocol with a chance-anchored, significance-tested discrimination skill. Deep backbones perform best on single-axis and harder multi-axis protocols. However, a classical histogram-of-oriented-gradients (HOG) descriptor trails the best frozen-network readout by a small (but significant) margin, is not statistically separable from the runner-up (a CNN-filter measure), and runs ~300× faster on CPU. Our results show that discrimination concentrates in mid-scale oriented features, where deep backbones peak at a low or mid stage, and HOG peaks at a mid cell size. Among existing methods, frozen deep features thus offer little over a tuned classical descriptor for measuring symmetry; whether task-trained deep scorers can do better remains open. We release the scorers and harness, in imgsym, an open toolkit for image symmetry detection and measurement. Full article

Adverse weather poses a major challenge to ground-based electro-optical (EO) aircraft surveillance. Existing parameter-efficient fine-tuning (PEFT) methods operate mainly in the spatial domain and treat weather as a generic domain shift rather than a frequency-dependent degradation. We propose Freq-LoRA, a frequency-domain PEFT method that applies the Type II Discrete Cosine Transform (DCT-II) to frozen encoder features, decomposes them into $K=4$ learned Gaussian frequency bands, and modulates the bands with an image-driven spectral gate. SpectralGate contains 140 parameters and estimates band importance from the input image’s DCT statistics, removing the need for external weather metadata at inference; weather-diverse training data are still required to learn the decomposition. On a Blender-simulated aircraft dataset covering five weather conditions, Freq-LoRA achieved a test mean Intersection-over-Union (mIoU) of 0.904, with a 95% confidence interval (CI) of [0.899, 0.908], using 559 K trainable parameters. Its point estimate differed by $-0.002$ from that of the weather-conditioned spatial method Feature-wise Linear Modulation (FiLM; 0.906), despite requiring no external weather metadata at inference. Relative to batch-size-matched Spatial LoRA (0.873; 95% CI: [0.867, 0.879]), Freq-LoRA had a 0.031 higher point estimate. Preliminary evaluation on real unmanned aerial vehicle (UAV) imagery yielded 0.421 mIoU (+13% relative to the zero-shot Segment Anything Model (SAM); one platform), and evaluation under six unseen image corruptions showed differences of at most 0.001 mIoU from the weather-oracle variant. Full article

Urban air mobility depends on small aerial vehicles threading through dense, wind-swept cities, yet most sampling-based planners treat the urban wind field as noise to reject rather than structure to exploit—and pay for it in flight energy. We take the opposite view. Behind every building lies a sheltered wake where the air slows and aerodynamic drag drops, and this work turns that physical fact into a planning principle. We present an energy-aware, wind-shadow-aware framework that routes a single quadrotor, at the planning level, through these low-wind corridors. The wind model couples a power-law shear profile with Ekman directional veer and a frozen-turbulence gust component, grounding the planner in realistic boundary-layer physics. A feed-forward neural energy surrogate, trained to approximate a cost field that aggregates wind exposure and obstacle clearance, then guides a two-stage refinement—energy-aware, collision-checked shortcutting followed by Laplacian and energy-guided smoothing—so that every accepted change stays collision-free. Against classical sampling-based baselines (RRT, goal-biased RRT, Informed RRT*, and BIT*) over a 50-run Monte-Carlo study, evaluated with multi-criteria metrics and Pareto-dominance analysis, the framework characterizes how wind-shadow-aware routing balances route energy against smoothness and clearance, offering a reproducible, wind-informed basis for flying robots navigation. Full article

Quantifying the soil moisture (SM) response to precipitation is pivotal for predicting hydrologic resilience and ecosystem stability in fragile cold regions. This is true in alpine permafrost environments characterized by variable precipitation phases and strong seasonal freeze–thaw dynamics. However, critical knowledge gaps persist due to the scarcity of high-resolution, multi-layer in situ observations in these remote areas. Using hourly data from three sites in the Tanggula Mountains (2020–2024), this study employs an event-based analytical framework combining logistic regression and linear regression to quantify multi-layer (10–100 cm) SM responses to rain, snow, and mixed-phase precipitation across seasons. Core findings indicate the following: (1) Precipitation thresholds with 80% probability of triggering SM responses rise sharply with depth during the cold period (10 cm: 1–11 mm; 50–100 cm: often >15 mm or unreachable) but increase gradually in the warm period (10 cm: 0.4–5 mm; 50 cm: <15 mm). Mixed-phase precipitation refers to the lowest amount of precipitation (0.4–2.5 mm at 10 cm), followed by rain (1–11 mm) and snow (2–5 mm). (2) Warm-period regression slopes are consistently steeper than cold-period slopes (at 10 cm, 0.0024 vs. 0.0010 for rainfall). Mixed-phase precipitation yields the steepest slopes, approximately 50% higher than rainfall at 10 cm in the warm period (0.0037 vs. 0.0024), due to its longer duration and dual-supply mode. For lag time, cold-period values are more widely dispersed due to multiple interacting factors, while warm-period values are concentrated; only warm-period rainfall exhibits a clear monotonic increase in lag time with depth, consistent with unsaturated flow theory. (3) The quantified regression slopes, threshold values, and phase-specific efficiencies provide transferable metrics for calibrating infiltration models and evaluating frozen-ground hydrology schemes. The finding that mixed-phase events are the primary driver of deep-layer recharge, despite accounting for a smaller fraction of the total event count, has direct implications for water resource assessment in high-altitude catchments where precipitation phase composition is often oversimplified. Overall, this study moves beyond qualitative descriptions by providing quantifiable, transferable metrics that advance the mechanistic understanding of precipitation–SM coupling in alpine permafrost regions. Full article

We introduce TriHex-Cluster, a hierarchical overlapping clustering framework built on the self-similar geometry of the triangular lattice (6-regular planar graph). The primary algorithm is regime C (greedy 2-packing followed by Voronoi completion), a practical hierarchical clustering method producing disjoint clusters with the Voronoi-contact graph $G_{\mathrm{Vor}}^{(k+1)}$ as the next-level graph and aggregation complexity $O(nlogn)$ (embedding cost excluded). On regular triangular domains with near-perfect packings, regime C achieves $n^{(k+1)}\le n^{\left(k\right)}/7+O\left(\sqrt{n^{\left(k\right)}}\right)$ per level; the measured depth on finite data is $K^{\ast }=\lceil {log}_{7}n\rceil \pm 1$. Two variants complete the framework: regime A (full-overlap edge-induced, ${\mathcal{C}}^{\left(k\right)}=V^{\left(k\right)}$) adds native overlap semantics by preserving the EI meta-graph 6-regularity without reducing the vertex count; regime B (deterministic index-7 sublattice, ${\mathcal{C}}^{\left(k\right)}={\mathsf{\Lambda }}_k$ with $\mathbf{a}=2-\omega$) is a theoretical construction establishing an exact sublattice density ratio of 7 per level on the infinite lattice $\mathbb{T}$, and exact termination in $K={log}_{7}n$ levels on finite periodic domains with $n=7^K$. Unconditional results: EI 6-regularity in regime A; perfect star-tiling and exact index-7 structure in regime B; strict hierarchy via Voronoi-completed clusters in regime C; tile-shape alternation proven at levels 1–2 (hexagonal, then triangular-like) and conjectured, with numerical verification, beyond; $h_{\mathrm{Ward}}$ (as an unnormalised SSE) strictly admissible and $h_{max}$ weakly admissible. Aggregation complexity, embedding excluded: $O(nlogn)$ in regime C, $O\left(n\right)$ in regime B, $O(n·K_{max})$ in regime A. We provide a fully reproducible reference implementation (trihex2, MIT-licensed) with extensive parameter sweeps on UCI benchmarks, synthetic Gaussians, non-convex shapes, and overlapping distributions. The genuine contributions of the framework are the multi-scale hierarchical structure with provable geometric guarantees and, in regime A, native overlap semantics that no hard-clustering baseline can provide. A central empirical finding concerns the embedding: an ablation isolating the 2D-lattice projection shows it to be the main bottleneck, and a lattice-free variant that runs the same combinatorial core directly on a k-nearest-neighbour graph in the original feature space—with no embedding and no quantisation—removes the projection entirely and improves accuracy on six of seven pilot datasets. With a frozen, fully unsupervised meta-selection rule (graph-geodesic arbitration between a convex-consensus and a graph-min-cut candidate, no per-dataset tuning), this variant reaches ARI $0.871$ on moons and $1.000$ on circles, where k-means, HAC, and GMM all collapse to $\le 0.43$ and $\le 0.00$, respectively. On a 73-dataset benchmark (23 real UCI, 50 synthetic, all loaded with validated class labels), TriHex is the most frequently best method on the synthetic panel ($46\%$ win rate) and close behind GMM overall ($34\%$ versus $36\%$), while having the lowest mean ARI—the signature of a specialist: it dominates on non-convex structure (rings, spirals, manifolds) and is outperformed on convex tabular data, where we make no claim of superiority. We also report a genuine robustness limitation: with the default configuration, TriHex fails on Cancer (ARI $0.042$, essentially uncorrelated with the ground truth) because the default lattice over-fragments a two-class problem; competitive performance requires a dataset-appropriate configuration, and we report this explicitly rather than only the best configurations. On overlapping Gaussians, regime A detects the boundary points that the data-generating process itself classifies as ambiguous with precision $1.00$ at heavy overlap ($\delta =0.5$); the detector over-flags as the clusters separate (precision falls to $0.43$ at $\delta =3.0$), so its usefulness is confined to the strong-overlap regime. Within that regime, it provides a measurable capability unavailable to hard-clustering baselines. Full article

Weakly supervised video anomaly detection (WS-VAD) localizes anomalous events in untrimmed videos using only video-level annotations. While CLIP-based methods have advanced this task through vision–language alignment, widely adopted approaches construct text prototypes from short category-name prompts of at most five words, leaving the CLIP text encoder not fully exploited. We propose SETAS-VAD, which addresses this gap through a Category Semantic Alignment (CSA) loss function: for each anomaly category, a large language model generates multi-sentence descriptions covering complementary semantic aspects, encoded once offline into frozen prototype vectors. An InfoNCE contrastive objective pulls attention-weighted anomaly features toward ground-truth category prototypes at zero additional inference overhead (prototype generation and encoding are performed once offline as a preprocessing step, not at test time). Under fully reproducible conditions on UCF-Crime and XD-Violence, SETAS-VAD achieves state-of-the-art temporal localization (30.45% mAP on XD-Violence, 12.16% on UCF-Crime), with per-threshold gains increasing at stricter IoU values, indicating improved boundary precision rather than coarse detection sensitivity. Full article

Sudden rupture of freezing pipes frequently occurs in artificial ground freezing engineering. Owing to complex field environments, it is difficult to clarify the single-factor evolutionary laws of vibration signals induced by pipe fracture through field monitoring. Four groups of model freezing pipes with diameter-to-thickness ratios ranging from 20 to 26 were fabricated through geometric scaling based on engineering prototype pipes of Φ140 × (5–7) and Φ159 × (6–8). A low-temperature brine medium at −30 °C was circulated inside the pipes to simulate actual in situ refrigeration conditions. A series of tensile rupture tests were performed to explore the vibration response characteristics of freezing pipes with different specifications. The test results indicate that the ultimate rupture load is positively correlated with signal energy. Within the diameter-to-thickness ratio range of 20–26, the signal amplitude decreases approximately linearly (R² = 0.99), while the progress count and signal energy increase gradually, and the dominant frequency of rupture vibration signals decreases continuously. In other words, the dominant frequency gradually declines as the diameter-to-thickness ratio rises. Unlike conventional acoustic emission characteristics of bare steel fracture, which are typically characterized by high-amplitude, high-energy bursts with prominent central frequencies, the vibration signals of freezing pipes surrounded by frozen soil exhibit significantly lower dominant frequencies (3–30 kHz), rapid attenuation due to pipe–soil energy radiation, and a frequency shift that is structurally governed by the diameter-to-thickness ratio rather than by material properties alone. The quantitatively established frequency bands and parameter evolution patterns can serve as reference criteria for field monitoring and early warning of freezing pipe fracture in artificial ground freezing engineering. Full article