Search Results (9,830)

Multimodal fusion unlocks the potential of point cloud semantic segmentation, thereby driving advancements in surface observation and visual perception tasks. Although light detection and ranging (LiDAR) systems capture precise 3D structural geometry and optical images provide rich semantic and textural information, existing fusion methods struggle with limited cross-modal perception and insufficient information complementarity. To address these limitations, we propose a multi-stage LiDAR-image collaborative perception fusion network (MCPFNet) for point cloud semantic segmentation of urban scenes. At the middle fusion stage, the network incorporates an elevation-guided geometric-aware fusion module and a semantic-aware cross-attention fusion module to enable bidirectional feature injection between LiDAR and image modalities. In the late fusion stage, a bidirectional adaptive fusion module further refines semantic representations through gated weighting and bidirectional cross-attention mechanisms. Extensive experiments on three multimodal datasets with different resolutions, i.e., ISPRS Vaihingen, N3C-California, and UAVScenes, demonstrate that MCPFNet outperforms existing fusion methods, achieving mIoUs of 74.51%, 95.15%, and 62.76%, respectively. Hence, our multi-stage fusion and bidirectional interaction strategy is more reliable and accurate than existing methods in performing segmentation across diverse and complex urban scenes. Full article

Depth super-resolution reconstructs high-resolution (HR) depth maps from low-resolution (LR) inputs with the aid of HR RGB guidance, but RGB edges often do not coincide with true depth discontinuities, causing texture copying and degraded geometric consistency. To address this problem, we propose Frequency–Geometry-Guided Network (FGGNet), a spatial–frequency fusion framework for RGB-guided depth map super-resolution. FGGNet introduces Multi-branch RGB-guided Convolution (MRGConv) to enhance RGB structural representations, a Geometry Prior-guided Fusion Module (GPFM) to filter geometrically inconsistent RGB responses using depth-derived priors, and radial complex spectral loss (RCSL) to emphasize boundary-related high-frequency components in the complex spectral domain. Experiments on NYU v2, Middlebury, Lu, and RGB-D-D show that FGGNet achieves competitive or superior reconstruction accuracy under synthetic and real-world degradation settings. Under the ×16 setting, FGGNet reduces RMSE by 13.7%, 22.8%, 18.5%, and 11.4% on the four datasets, respectively, compared with the average RMSE of five representative state-of-the-art methods. These results validate the effectiveness of combining geometric prior filtering with frequency-domain supervision for reliable depth reconstruction. Full article

The electrocatalytic nitrate reduction reaction (NO₃RR) enables sustainable ammonia synthesis from nitrate waste, yet its complex mechanism and severe competition from the hydrogen evolution reaction (HER) demand precise control over interfacial electronic structures. This review provides a mechanistic overview of interfacial electron engineering for NO₃RR via charge transfer, d-band center modulation, and d-p orbital coupling. We propose a reverse-engineering framework that starts from the three kinetic bottlenecks of NO₃RR (nitrate activation, *H supply, and intermediate poisoning) and back-extracts the required electronic effects (charge transfer, d-band shift, and d-p orbital coupling). From this perspective, we cover the construction of built-in electric fields (BIEFs) in heterojunctions, engineering atomic-scale active sites (e.g., single-atom and dual-atom catalysts), and exploiting hydrogen spillover and reverse spillover for cross-spatial proton delivery. Given that rational interfaces dynamically evolve under operating conditions, we highlight that in situ/operando characterization captures the dynamic restructuring of valence states, coordination environments, and morphologies, establishing clear structure–electron–activity relationships. Finally, we discuss key challenges and outline future directions, including machine learning-accelerated screening, dynamic interface regulation, and synergistic integration of multiple electronic effects. This review offers a comprehensive framework for interfacial electron engineering, guiding rational design of next-generation NO₃RR electrocatalysts. Full article

Variational autoencoders (VAEs) are sensitive to spatially heterogeneous noise because corrupted high-frequency components can be propagated into the latent posterior and degrade reconstruction. We propose Learnable Local FFT-VAE, a VAE-oriented localized frequency correction framework that performs windowed Fourier-domain modulation before variational encoding. The proposed Localized Denoising-Correction Module extracts overlapping patches, applies local FFT, predicts patch-conditioned frequency-control parameters $(\alpha ,\sigma ,m)$, and reconstructs corrected inputs through inverse FFT and Hann overlap-add. This design preserves spatial locality while remaining differentiable and jointly trainable with the VAE objective. We evaluate the method on FashionMNIST, CIFAR-10, and PneumoniaMNIST under homogeneous and heterogeneous synthetic noise, and compare it with Direct VAE, AE, Local FFT-AE, best-tuned Fixed FFT-VAE, learnable Global FFT-VAE, BM3D, DnCNN, and FFDNet. The revised experiments demonstrate improvements in PSNR, SSIM, SNR, MSE, HFEN, and gradient-error, with statistical tests over random seeds. Parameter heatmaps and spectral visualizations further show that the learned local correction adapts to spatially varying corruption. PneumoniaMNIST is used only as a lightweight biomedical benchmark, and the method is not intended for clinical diagnosis. Full article

Ulomoides dermestoides (Yangchong) is a tenebrionid beetle used in traditional medicine across Asia and Latin America. While crude extracts show effects on inflammation, oxidative stress, and other conditions, systematic integration of its bioactive substances, mechanisms, and translational potential is lacking. This review consolidates its chemical basis, comprising volatile benzoquinones, terpenes, and alkenes, alongside non-volatile fatty acids, proteins (antioxidant enzymes, glycoproteins), and phenolics. Pharmacological evidence indicates multi-target modulation of reactive oxygen species (ROS), cytokines, leukocyte recruitment, endothelial activation, and thromboinflammation. Recent advances include proteomic identification of antioxidant protein complexes, neuroprotection in a Parkinson’s disease model, chromosome-level genome assembly, and isolation of the UDP-glucose pyrophosphorylase 2a (UGP2A) glycoprotein, which alleviates thrombosis partly via toll-like receptor 4/myeloid differentiation primary response 88 (TLR4/MyD88)-mediated endothelial anti-inflammatory effects. However, most evidence remains preclinical, relying on non-standardized crude extracts, and benzoquinone-containing fractions display potential cytotoxicity and genotoxicity. Future research should integrate bioassay-guided isolation, structural characterization, multi-omics, pharmacokinetic/pharmacodynamic (PK/PD) analysis, standardized quality markers, and rigorous safety evaluation to transform U. dermestoides from an empirical insect-derived medicinal resource into a scientifically validated source of preclinical antioxidant and anti-inflammatory candidate substances. Full article

Platform motion errors degrade SAR image quality in terms of severe defocusing and azimuth blurring. We propose a Range-aware Autofocus Network (RAFnet) by embedding a novel range-aware attention module into a progressive autofocus framework. The module exploits 1-D azimuth pooling to compress spatial features and extract high-SNR scattering components from the range dimension. Such features are further enriched via a light cross-channel interaction. To facilitate coarse-to-fine hierarchical learning, we develop a progressive multi-scale entropy loss which jointly optimizes the entire network. Experimental results on real SAR data show that the proposed approach captures high-level phase fluctuations accurately and effectively suppresses raw phase deviations and sidelobes. Quantitative results show that combining the attention module with multi-scale loss achieved a global spatial entropy of 9.8567 and contrast of 5.0091 in focused images. By extracting more accurate focus-oriented feature representations, we provide an effective solution for high-quality SAR auto-focusing. Full article

In Unmanned aerial vehicle (UAV) forest fire detection, flames and smoke exhibit dramatic scale variations. Existing methods often struggle with multi-scale feature extraction, fusion quality, and localization reliability, resulting in limited accuracy improvements. To address this issue, this study optimizes the backbone, neck, and head of YOLOv11n to propose a novel multi-scale forest fire detector (MFFDet), which consists of three key modules: (1) the Multi-Scale Feature Calibration Module (MFCM) is designed to improve multi-scale feature representation by context aggregation and detail calibration; (2) the Cross-Scale Semantic Alignment Module (CSAM) is proposed to enhance fusion quality by applying channel reorganization and local spatial refinement; and (3) the Location Quality Estimator Head (LQEH) is presented for reliable localization by mapping the statistical information of regression distributions into a localization quality score, which systematically boosts the accuracy and stability of multi-scale object detection. In addition, to alleviate the scarcity of UAV forest fire detection data, this study constructs a UAV Forest Fire Dataset (UF${}^2$D), providing important data support for UAV-based fire detection. Experiments on UF${}^2$D show that MFFDet achieves an mAP@0.5 of 70.1%, the best among all compared models, representing a 4.4% improvement over the baseline. Moreover, it attains the top performance on small, medium, and large objects, with AP${}_s$ of 20.3%, AP${}_m$ of 31.5%, and AP${}_l$ of 44.8%, highlighting MFFDet’s robustness and accuracy for multi-scale flame and smoke detection in a complex forest fire environment, which bears important practical significance for the intelligent upgrade of forest fire prevention and control. Full article

Critical illness myopathy (CIM) is linked to mechanical ventilation and complete mechanical muscle silencing in intensive care unit (ICU) patients. Limb muscles show atrophy and declined specific single fibre force through altered protein turnover and diminished myosin-to-actin ratios. A rat ICU model reproducing preferential myosin loss and specific force decline in limb muscle was used to assess myofibrillar remodelling. After 5 or 10 days of ICU intervention, single extensor digitorum longus (EDL) and soleus muscle fibres were imaged using label-free, SecondHarmonic Generation (SHG) microscopy, followed by quantitative 3D morphometry. The degree and severity of deranged myofibrillar architecture was assessed through (i) 2D and 3D Cosine Angle Sums (CAS2D/CAS3D) and (ii), Vernier Densities (VD) parameters. A progressively declining myofibrillar order was seen by dropping CAS2D/3D and increasing VD values during ICU intervention, reflecting angular and axial register deviations. Effects of chaperone co-inducer BGP-15, dissociative glucocorticoid Vamorolone (VBP-15) or its parent compound Prednisolone on myofibrillar architecture were explored. Both BGP-15 and VBP-15 modulated the progression of myofibrillar disorder seen during ICU intervention alone: For soleus fibres, BGP-15 maintained structural integrity at day 5 but not at day 10, while even worsening myofibrillar order in the EDL. VBP-15 reversed atrophy at day 10 in soleus but not in EDL fibres. Our study is the first to quantify myofibrillar remodelling in limb muscle fibres during ICU intervention in 3D and provides exploratory assessment of BGP-15 and VBP-15 treatments on aberrant remodelling in CIM. Full article

Background: Oxidative stress and chronic inflammation contribute to the pathogenesis of obesity and type 2 diabetes (T2D), yet the effects of dietary interventions on endogenous antioxidant defences remain poorly defined. This is a non-randomized study evaluates the effects of very-low-energy ketogenic therapy (VLEKT), compared with a Mediterranean diet (MedD) and a control group, on antioxidants, metabolic, and inflammatory markers. Materials and Methods: Thirty adults with obesity and T2D were assigned to VLEKT (n = 10), MedD (n = 10), or control (n = 10) for 90 days. Metabolic parameters, inflammatory cytokines, superoxide dismutase (SOD) and glutathione peroxidase (GPx) activities were assessed. Longitudinal changes were analyzed using linear mixed models. Results: VLEKT exhibited significant reductions in body weight, fat mass, HbA1c, and HOMA-IR. SOD activity increased in the VLEKT group, whereas no significant changes were observed in MedD. Changes in SOD were inversely associated with changes in HOMA-IR. GPx showed a less consistent response pattern, while inflammatory markers did not differ between groups. Conclusions: VLEKT was associated with substantial metabolic improvement accompanied by a selective modulation of antioxidant enzyme activity. The increase in SOD activity and its association with HOMA-IR suggest a link between metabolic and redox adaptations in subjects with obesity and T2D. Full article

This paper proposes a hybrid Vector Radiative Transfer Shooting (VRT-S)-Bouncing Ray (BR) method, referred to as the VRT-S-BR method, for predicting composite electromagnetic scattering from targets above vegetation-covered rough surfaces. In this proposed framework, the vegetation layer is modeled as a stratified random medium and incorporated into the BR solver through VRT-S-derived amplitude modulation and deterministic phase compensation. Specifically, an offline database of vegetation-induced complex reflection coefficients is first generated using the VRT-S model over a set of incidence angles. During the BR ray-tracing process, these coefficients are used to replace the conventional Fresnel reflection terms on a per-interaction basis, thereby accounting for vegetation-induced attenuation and coherent scattering effects. In addition, a facet-dependent phase compensation scheme is introduced to describe propagation-path variations of individual rays through the vegetation canopy, avoiding the empirical random phase perturbation used in previous hybrid models. The proposed method is validated against field-measured backscattering data over natural grassland, achieving root mean square height (RMSE) values of 1.82 dB and 3.10 dB for horizontal-horizontal (HH) and vertical-vertical (VV) polarizations, respectively. Numerical results further demonstrate the capability of the method to characterize target–vegetation coupled scattering under different percentages of vegetation cover, vegetation heights, terrain backgrounds, and bistatic observation geometries. Full article

Postharvest quality deterioration of Korla fragrant pear (Pyrus sinkiangensis Yu) severely constrains its market value, yet the regulatory role of preharvest soil management in shaping postharvest performance remains poorly understood. Although green manure is widely adopted to ameliorate orchard soil degradation, species-specific modulation of postharvest storage trajectories and the quantitative fidelity of soil-to-fruit nutrient transmission have rarely been resolved for climacteric pear species. This study investigated how green manure species modulate fruit quality at harvest and during postharvest storage life and their underlying soil–fruit linkages. Three preharvest treatments were imposed, as follows: control (CK), sweet clover (CM), and alfalfa (MX). Fruits were harvested and stored at 4 °C, with samplings at 1, 5, 10, 15, and 20 d. A critical quality transition was identified at 15 d, characterized by the concurrent peaking of soluble sugars, organic acids, vitamin C, and anthocyanins alongside an optimal sugar–acid ratio. Beyond this inflection point, CM and MX diverged markedly: CM enhanced soluble sugar accumulation, anthocyanin retention, and ester volatile production—most notably hexyl acetate, which increased over 14.4-fold—thereby generating a pronounced fruity aroma bouquet. Conversely, MX sustained higher amino acid and vitamin C levels and conferred superior late-storage stability, evidenced by a three-fold lower coefficient of variation in the sugar–acid ratio relative to CK. Partial-least-squares structural equation modeling (PLS–SEM) revealed soil fertility as the principal exploratory associative factor of fruit quality, but the fidelity of soil-to-fruit transmission was species-dependent. MX exhibited the highest observed associative strength (R² = 0.971), whereas CM exhibited attenuated transmission fidelity (R² = 0.777), with network analysis further indicating that CM exhibited divergent associative patterns of key soil–fruit correlations. These findings suggest that green manure identity is linked to postharvest quality through divergent soil–fruit coupling pathways: alfalfa shows nutrient transmission efficiency and stabilizes nutritional quality, whereas sweet clover promotes sugar-aroma accumulation at the cost of reduced soil–fruit conversion fidelity. Species-specific green manure selection thus offers a viable strategy for targeted modulation of postharvest traits in Korla fragrant pear. Full article

When acquiring point cloud data of precision parts using 3D scanning devices, occlusion or equipment limitations often lead to sparse and incomplete data, resulting in the distortion or loss of key geometric features. To address this issue, this study proposes an attention-enhanced feature-based point cloud completion network for precision parts, using precision bearing rings as an example to construct a dedicated completion dataset for training. The proposed network adopts an encoder–decoder architecture. In the encoder stage, a curvature-weighted sampling feature extraction module and spatial attention mechanism are introduced to extract both local and global features from the incomplete point cloud, followed by multilevel feature fusion. The multiscale features extracted by the encoder are then fed into the decoder, which hierarchically and progressively predicts the missing regions of the point cloud. Finally, an adversarial generation module incorporating a biased attention mechanism enhances the sensitivity of the network to geometric structural differences, thereby producing a complete and refined point cloud as the final output. Experimental results show that on the ShapeNet-part dataset, the proposed network achieves average CD, Pred → GT, and GT → Pred errors of 4.663, 2.459, and 2.457, respectively, representing reductions of 10.8%, 4.7%, and 8.1%, respectively, compared with the mainstream PF-Net completion network. On the bearing ring dataset constructed in this study, the average CD, Pred → GT, and GT → Pred errors were 0.497, 1.064, and 0.601, respectively, decreasing by 9.3%, 16.3%, and 16.2%, respectively, relative to PF-Net. Moreover, the proposed network effectively completed the point clouds of various missing parts, demonstrating its robustness across different types of precision parts. Full article

Lithological mapping with multispectral remote sensing remains challenging when diagnostic spectral information is limited and reliable labeled samples are scarce. This problem is particularly relevant when convolutional neural networks (CNNs) are applied to lithological classification, because limited spectral dimensionality and scarce training samples may hinder the learning of discriminative spatial–spectral features. In this study, we developed a limited-sample lithological mapping framework for the Shibaocheng area of Subei County, Gansu Province, China, using band-integrated ASTER and Sentinel-2A multispectral imagery. ASTER shortwave infrared (SWIR) bands were co-registered and resampled to Sentinel-2A imagery, and then integrated with Sentinel-2A visible and near-infrared (VNIR) and red-edge bands to construct a complementary multispectral dataset. A compact spectrally enhanced multi-scale CNN was designed, incorporating a residual spectral feature enhancement module for inter-band representation learning and a parallel multi-scale hybrid convolution module for capturing spatial–spectral features. Eight lithological units were classified under limited-label conditions using 8158 training samples and 3497 spatially independent validation samples. Experimental results show that the band-integrated ASTER–Sentinel-2A dataset improved classification performance compared with single-sensor inputs. Using the proposed model, the band-integrated dataset achieved an overall accuracy (OA) of 94.12%, average accuracy (AA) of 94.04%, and Kappa coefficient of 0.932, compared with OA values of 93.14% and 92.40% obtained using ASTER and Sentinel-2A alone, respectively. The positive effect of band-level integration was also observed for spectral angle mapper (SAM), support vector machine (SVM), and 3D-CNN, whose OA values increased to 54.33%, 86.12%, and 92.29%, respectively. The proposed CNN achieved the highest OA among the evaluated methods, outperforming SAM, SVM, and the conventional 3D-CNN. In addition, t-SNE visualization indicated that incorporating spatial texture features produced more compact and better-separated lithological clusters than using spectral features alone. Ablation experiments further demonstrated that the proposed spectral feature enhancement and multi-scale hybrid convolution modules each contributed to improving lithological classification performance. These results demonstrate that integrating freely available multispectral data with a lightweight spectral–spatial CNN provides a practical and cost-effective solution for lithological mapping in bedrock-exposed arid to semi-arid regions, especially where hyperspectral imagery and dense field samples are unavailable. Full article

Aiming at core bottlenecks in weak and small target detection in radar range-Doppler (RD) spectra under low signal-to-noise ratio (SNR)—including severe performance degradation of traditional constant false alarm rate (CFAR) detectors and the inherent trade-off difficulty faced by existing deep learning methods in balancing detection accuracy, localization precision, and real-time performance—this paper proposes a progressive sub-pixel-level intelligent detection algorithm named F²DN-CCWL. The algorithm constructs a three-stage detection pipeline: global candidate screening, local fine discrimination, and weighted localization, and implements a full-stack customized design covering network architecture, soft-label training strategy, and post-processing modules. Simulation and field-measured results demonstrate that at −20 dB SNR, the proposed algorithm achieves a detection probability of 95.3%, a false alarm rate of 3.1%, an average localization error of 0.76 pixels, and a single-frame inference latency of 47.21 ms. This method offers a high-performance engineering solution for radar-based detection of low observable targets. Full article

The autism spectrum disorder-associated SCN2A, ANK2, and SHANK-family genes encode molecularly distinct proteins that converge functionally on dendritic integration. Recent work established that ankyrin-B, encoded by ANK2, acts as an obligate dendritic scaffold for NaV1.2, encoded by SCN2A, in neocortical pyramidal neurons. Loss of this module mislocalizes dendritic NaV1.2, reduces dendritic Na⁺ influx, weakens backpropagating action potentials, and impairs synaptic maturation and long-term potentiation. SHANK proteins organize a complementary postsynaptic receptor scaffold within dendritic spines, coupling N-methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and metabotropic glutamate receptor (e.g., mGluR5) signaling to the actin cytoskeleton through layered PSD-95/GKAP/Homer interactions. Disruption of this scaffold can destabilize excitatory transmission, spine morphology, and plasticity. We propose that these dendritic shaft and spine-associated modules jointly regulate dendritic input–output gain and that their disruption may contribute to autism spectrum disorder by destabilizing, rather than uniformly shifting, excitatory integration across cortico-subcortical circuits relevant to sensory reactivity, behavioral flexibility, and social-valence processing. Here, we review the cortical evidence for this layered dendritic convergence and evaluate its potential relevance beyond the cortex. We assess the striatum, thalamus, and amygdala as subcortical sites where related dendritic scaffolding mechanisms may operate. The striatum provides the strongest current test case, with established roles for both NaV1.2 and SHANK3 in medium spiny neuron physiology and corticostriatal connectivity. Thalamic and amygdalar extensions are supported mainly by SHANK-related circuit and channelopathy data but lack direct evidence for ANK2–SCN2A involvement. The framework is experimentally testable: conditional Ank2 deletion in striatal, thalamic, and amygdalar cell types; dendritic Na⁺/Ca²⁺ imaging across Scn2a, Ank2, and Shank3 models; adult rescue experiments; and genetic-interaction designs would determine whether ankyrin-B supports dendritic excitability beyond the cortex and whether these genes converge on, rather than merely parallel, dendritic input–output gain. Validation in human subcortical tissue would then establish whether this dendritic scaffolding logic represents a shared point of convergence through which genetically distinct autism spectrum disorder-risk variants alter circuit function. Full article