Search Results (654)

Scene graph generation (SGG) provides a structured representation for visual understanding. However, most existing methods are designed to optimize global triplet recall rather than retrieve relation instances specified by a user query. In query-driven visual relation retrieval, two major challenges arise: the target relation must compete with a highly redundant candidate space, and query semantics are not incorporated before relation classification. To address these challenges, we propose a Query-Driven Candidate Relation Screening (QCRS) module, which injects query semantics into the candidate screening process. Specifically, QCRS encodes the query and candidate visual relation features, and then filters query-relevant candidates through relevance scoring. By reducing interference from irrelevant candidates and avoiding redundant computation, QCRS improves the final exact triplet hit performance and enhances the interpretability of query-specific relations, thereby facilitating query-driven visual relation retrieval. Built upon the strong EGTR baseline, QCRS learns query relevance to prioritize relation instances matching the target query, enabling precise triplet retrieval. Extensive ablation studies and analyses on the VG150 benchmark validate the effectiveness of the proposed approach: when integrated with EGTR, QCRS improves PairR@50 from 61.52% to 80.06% and ETR@50 from 30.54% to 47.07%, achieving absolute gains of over 16 percentage points in both correct object pair retention and end-to-end target relation retrieval performance. Full article

Network technology has advanced rapidly in recent years, and distributed denial-of-service (DDoS) attacks have grown more diverse, stealthy, and large-scale. Traditional detection approaches struggle to process long network traffic sequences and locate sparse attack signals hidden in massive normal traffic, which makes accurate and efficient DDoS detection an urgent requirement. This paper presents an end-to-end DDoS detection model built on the Mamba architecture. We use one-dimensional convolutions to extract local features and smooth noise, which strengthens the model’s ability to capture bursty attack behaviors. Then, taking advantage of Mamba’s linear complexity and selective scanning mechanism, the model models long traffic sequences, filters out redundant information, and concentrates on potential attack patterns. With global feature aggregation and a classification layer, the model realizes accurate attack recognition. Experiments conducted on the CIC-DDoS2019 dataset show that our model obtains better performance in weighted F1 score, precision, and recall, while also improving inference efficiency. The model is suitable for high-precision, low-latency DDoS detection in real network environments. Full article

Addressing the challenges of high-dimensional redundancy, noise interference, and parameter missing in the net present value prediction of shale gas wells, an intelligent prediction model PCA-DFNN integrating Principal Component Analysis and Deep Feedforward Neural Network is proposed. Based on actual data from 48 shale gas wells, Principal Component Analysis is first performed on 19 input features to reduce dimensionality, extracting 9 core principal components, which achieve a cumulative variance contribution rate of 88.05%. Subsequently, a deep neural network model is constructed for comparative modeling. The results indicate that the PCA-DFNN model achieves a coefficient of determination on the independent test set that improves from 0.6439 in the original model to 0.6882, an increase of 0.0443, or approximately 6.9%, with faster training convergence and superior generalization ability. The research confirms that the proposed method can effectively eliminate feature redundancy, filter noise, and circumvent the uncertainty of missing value imputation, providing a more reliable technical tool for the early economic evaluation of shale gas. Full article

User behaviors and social interactions on online platforms are intricately intertwined, naturally forming complex graph structures. Leveraging this structure, Graph Neural Networks (GNNs) efficiently aggregate neighborhood information and have become a prevailing paradigm for social recommendation. However, existing methods often overemphasize social modeling while overlooking the joint effects of preference-guided relation filtering and user/item biases, rendering them vulnerable to noise from redundant ties. To address these limitations, we propose PDDSR, a Preference-Guided Debiasing and Denoising Social Recommendation framework. Specifically, for debiasing, PDDSR explicitly models user rating bias and item popularity bias as learnable vectors, integrating them into embedding learning to mitigate bias drift at the embedding level. Simultaneously, for denoising, the model employs a social relation confidence mechanism guided by user preferences and adopts an adaptive graph denoising strategy to retain highly informative connections, effectively capturing social influence while filtering out noise. Extensive experiments on the Ciao and Epinions datasets demonstrate that PDDSR consistently outperforms state-of-the-art methods, and notably on the Ciao dataset, the MAE and RMSE are improved by 1.90% and 1.87%, respectively. These results validate the effectiveness and robustness of the joint debiasing and denoising mechanism in complex social recommendation scenarios. Full article

We propose NPMV-SLAM, an RGB-Depth neural implicit dense SLAM system based on multi-resolution voxels and feature point clouds. The method demonstrates exceptional performance in tracking accuracy and reconstruction quality, aiming to address the shortcomings of traditional visual SLAM in texture detail modeling and geometric consistency, as well as the limitations of existing neural implicit methods in real-time performance and scene scalability. (1) We innovatively propose a position-enhanced encoding mechanism that fuses multi-resolution hash voxel grids with feature point clouds. This design fully leverages the high sensitivity of point clouds to high-frequency geometric details and the global structural continuity provided by voxels, achieving complementary advantages during network training and inference, thereby comprehensively enhancing the system’s reconstruction generalization capability. (2) Furthermore, we design an adaptive sampling strategy guided by point cloud density priors. This strategy fundamentally alleviates the core issue of insufficient scene scalability through data-driven online point cloud reconstruction. By filtering out invalid, non-surface sampling points, it concentrates computational resources on object surface regions, significantly reducing computational redundancy in empty areas, and achieves efficient point cloud spatial indexing with the aid of a vector database similarity search algorithm. While maintaining operational efficiency, our method significantly improves both detailed reconstruction capability and global reconstruction completeness. Experiments conducted on multiple indoor scenes from the Replica and TUM datasets show that our approach achieves notable improvements in tracking accuracy, rendering quality, and mapping accuracy, successfully balancing precision and efficiency. Full article

Visual-Inertial Odometry (VIO) estimates system pose by fusing visual and inertial measurements. Although line features can enhance structural perception, existing approaches still face challenges such as redundant short segments and weak geometric constraints. To address these, in the front end, we propose a complete geometric optimization pipeline for line features. This pipeline adopts a length-threshold-based filtering strategy and integrates the proposed geometric-consistency-based merging mechanism, endpoint-distance-based verification mechanism, and epipolar-constraint-based triangulation method, transforming fragmented short segments into structurally complete 3D spatial lines. In the back end, reprojection residuals of the optimized line features are jointly optimized with point residuals, IMU pre-integration residuals, and marginalization priors in a sliding-window framework. Experiments on the EuRoC dataset show that compared to VINS-Mono, PL-VINS, and EPLF-VINS, the proposed method reduces the Absolute Pose Error (APE) by 17.57%, 9.88%, and 6.65%, respectively. Additionally, compared to PL-VINS, it reduces the line feature processing time by 4.16% and the average per-frame processing time by 2.36%, validating the effectiveness of the proposed method. Full article

High-resolution panoramas generated by UAV image stitching are indispensable image resources for remote sensing applications. However, most existing stitching methods are designed for small-size images, making it difficult to process large-size images efficiently, leading to problems such as image feature misalignment and low generation efficiency. This paper presents LargeStitch, a novel batch stitching method for large-size UAV images. The method introduces advanced image matching and alignment strategies through deep learning techniques to achieve efficient extraction and accurate alignment of dense features. To further optimize the stitching effect, this paper also proposes a seamless fusion method based on Seam-band, which effectively solves the problem of ghosting and misalignment in the overlapping region of large-size images. In addition, we designed a mask-based pre-stitching image filtering strategy, which optimizes the selection of candidate images to reduce content redundancy, thereby effectively avoiding unnecessary computational overhead and time consumption. The experimental results show that LargeStitch is not only capable of realizing fast stitching of high-precision and large-size aerial images but also significantly outperforms existing methods in terms of stitching quality and processing efficiency, making it a practical solution for realizing high-efficiency and seamless aerial image stitching. Full article

Corner detection is a fundamental task in computer vision that plays a critical role in applications such as image registration, 3D reconstruction, and object tracking. In biological visual systems, simple cells in the primary visual cortex exhibit high selectivity to edge stimuli of specific orientations, while end-stopped cells can detect geometric singular structures such as line segment endpoints and corners. Existing corner detection methods based on visual neural computation typically employ a strategy of densely distributed end-stopped cells for corner localization, which suffers from significant localization deviation under small angle conditions due to mutual interference between responses of adjacent neurons. To address this problem, this paper proposes a bionic corner detection method based on cooperative processing of simple cells and end-stopped cells. The method constructs a two-stage cooperative processing framework: the edge filtering stage employs a Gabor filter bank to simulate the orientation selectivity of simple cells, extracting edge positions and orientation information; the dynamic construction stage builds unilateral end-stopped cells only at filtered edge positions based on local orientation information, fundamentally avoiding computational redundancy and response interference caused by global dense distribution; the corner localization stage determines precise corner coordinates through hierarchical clustering and dual-cluster centroid fusion strategies. Experimental results demonstrate that, in the 15° acute-angle regime where dense end-stopped schemes are most severely affected by response interference, the proposed method reduces the mean localization error from 8.76 to 2.34 pixels, corresponding to a 73.3% improvement; averaged across the eight tested angle levels from 15° to 165°, the improvement is approximately 40.9%, and all per-angle differences are statistically significant (paired t-test, p < 0.01 or below, N = 10 independent runs). On standard test images, the method attains the lowest mean localization error among the eight compared detectors (1.58 pixels, versus 1.68–3.42 pixels for Harris, FAST, COSFIRE, KAZE, SuperPoint, Deep Corner, and Wei et al.), while maintaining competitive detection rate, false-alarm rate, and runtime. Physiological plausibility validation experiments show that the correlation coefficient between the detection deviation of this method and human perceptual deviation reaches 0.923, indicating that the output of the framework aligns with previously reported human perceptual bias patterns and supporting its biological plausibility as a biologically inspired—rather than mechanistic—model of corner perception. The source code, dataset, and experimental results are publicly available (see Data Availability Statement). Full article

Relation Extraction is crucial for knowledge graph construction, but extracting complex relations in specialized domains like Satellite Navigation Countermeasures (SNCM) remains challenging due to long semantic spans and high relational density. While Large Language Models (LLMs) possess strong semantic understanding, they often suffer from severe recall deficiency and hallucinations in high-density multi-entity contexts. Conversely, traditional small models generate excessive redundant noise. To address these limitations, this paper proposes an evaluation-model-guided relation extraction method (E-guidedRE). This framework employs a two-stage collaborative mechanism. First, a lightweight evaluation model utilizing a GlobalPointer network with Rotary Position Embedding (RoPE) and a sparse multi-label loss function acts as a structural filter to generate high-coverage candidate entity pairs. Second, these candidates guide the frozen LLM to perform deep semantic discrimination and retrospective denoising. Furthermore, we construct a dedicated SNCM dataset to fill the vertical domain data void. Extensive experiments across five heterogeneous datasets, including general, biomedical, financial, and our self-built SNCM corpus, demonstrate that E-guidedRE exhibits remarkable robustness. In ablation studies on the SNCM dataset, our method improved the F1-score from 36.54% to 54.93% compared to standalone LLM extraction, boosting recall from 27.81% to 47.13%. The proposed paradigm effectively mitigates the LLM’s attention divergence in complex contexts, dynamically balancing precision and recall, and offers a highly reliable technical pathway for knowledge extraction in specialized vertical domains. Full article

Wetland ecosystems have been a central focus of ecological research for an quite some time. Nevertheless, the degradation of wetland riparian zones has markedly accelerated due to anthropogenic activities, climate change, and habitat heterogeneity. The objective of this paper is to investigate the differences in functional traits of riparian plants under changing wetland environments on a karst plateau, as well as to elucidate the adaptive strategies of wetland plants across different habitats. This study examines the Caohai Wetland located on the Guizhou karst plateau, selecting the leaves of four dominant plant species (Phragmites australis, Onopordum acanthium, Galium odoratum, Paspalum distichum) in the Caohai Wetland lakeshore zone and analyzes the influence of soil factors on the variation of plant functional traits within the wetland riparian zone. The results reveal that: (1) significant differences exist in the functional traits of dominant plants in the riparian zones of karst plateau wetlands, with complex interrelationships among these traits; (2) the coefficients of variation for magnesium (Mg) and calcium (Ca) in the soil are notably high (79.53% and 67.21%, respectively), whereas soil oxidation-reduction potential (ORP) exhibits the lowest coefficient of variation (4.36%)—furthermore, the convergent variation in specific leaf area (SLA) and leaf dry matter content (LDMC) directly reflects the strong environmental filtering imposed by this habitat—and (3) redundancy analysis (RDA) indicates that leaf length (LL), specific leaf area (SLA), leaf area (LA), and plant carbon content (PCC) are particularly sensitive to environmental changes, while soil calcium (Ca), total nitrogen (TN), water-dispersible clay (WDR), soil organic matter (SOM), soil moisture content (SPMC), and total potassium (TK) constitute the principal soil factors influencing plant adaptive strategies in karst plateau wetlands. In conclusion, this study demonstrates that adaptation to karst wetland habitats is mediated through trade-offs in the allocation of photosynthetic products and the regulation of carbon (C), nitrogen (N), and phosphorus (P) nutrient balances under calcium-enriched and phosphorus-limited conditions, thereby reflecting the response characteristics of functional traits in karst plateau wetland plants to environmental changes. Full article

In robotic grinding of complex curved surfaces, the low stiffness of serial robots causes tool tip deflection and degrades surface quality. The axial symmetry of grinding discs introduces a free rotational parameter at each waypoint, converting a standard 6-DOF robot into a functionally redundant system. However, this redundancy has not been systematically exploited for stiffness optimization along the trajectory. This paper proposes a hierarchical dynamic programming framework to optimize the redundancy angle sequence over the entire grinding trajectory. A kinematic transformation parameterizes the flange target by the redundancy angle, enabling enumeration of feasible candidate configurations over a discretized grid. A composite stiffness index that accounts for the normal, feed, and cross-feed grinding force components is formulated at the contact point. Hierarchical constraint filtering removes configurations that violate posture, singularity, velocity, acceleration, and stiffness constraints. The Viterbi algorithm then recovers the minimum-cost path that balances stiffness performance and joint motion smoothness. Finally, a post-processing step based on a cubic smoothing spline generates $C^2$-continuous joint trajectories. Simulations on a UR5 robot grinding a curved surface evaluate the proposed framework against fixed-angle, greedy, and flange-stiffness baselines. The proposed method improves the mean composite stiffness by 31.7% and 17.9% over the fixed-angle and flange-stiffness baselines, respectively, and reduces the maximum joint jump by two orders of magnitude compared with the greedy strategy. Experimental validation on a UR5 robot confirms that the smoothed trajectory is accurately tracked while the stiffness threshold is preserved. A multi-trajectory analysis further shows that the stiffness threshold is maintained across all grinding trajectories. These results demonstrate the effectiveness of the proposed framework for redundancy optimization in robotic grinding with tool spin symmetry. Full article

The unchecked expansion of tea plantations onto steep, forest-adjacent slopes in subtropical mountains engenders a conflict between agricultural productivity and ecosystem integrity, particularly by exacerbating habitat fragmentation and soil erosion. While precise monitoring is essential to navigate this trade-off for sustainable management, accurate inventorying remains a challenge due to the plantations’ strong phenological variability, heterogeneous canopy structures, and high spectral confusion with surrounding vegetation. This study proposes a feature-optimized deep learning framework for mapping and characterizing tea plantations in complex landscapes, using Xinyang City, China, as a study area. The framework integrates multi-temporal Sentinel-1/2 observations with a sequential Jeffries-Matusita (JM)-Pearson feature filtering strategy. This approach effectively condenses a 132-variable high-dimensional pool (including optical spectra, vegetation indices, textures, and SAR polarimetry) into a compact 28-feature subset (a 78.8% reduction), preserving critical phenological and structural cues while minimizing redundancy. These optimized predictors drive a hybrid VGG16–UNet++ segmentation network, which couples transfer-learning-based semantic encoding with detail-preserving dense skip fusion. Extensive experiments across 18 model–feature configurations demonstrate that the optimal setting achieves an Overall Accuracy of 97.82%, an F1-score of 0.9093, and a mean IoU of 0.7968. Notably, the method significantly reduces misclassification in rugged, cloud-prone terrain, yielding a User’s Accuracy of 91.14% for tea. Based on the generated wall-to-wall map, we derived two decision-support indicators: multi-threshold steep-slope exposure and a normalized tea–forest interface density. This framework provides actionable, high-precision spatial products to support slope-based zoning, ecological restoration, and sustainable management in fragile mountain agroforestry systems. Full article

To reduce the conservatism of local trajectory planning in dynamic road scenarios caused by redundant projection of predicted trajectories, this paper proposes a hierarchical hybrid trajectory-planning framework with a time-to-arrival (TTA)-driven dynamic risk-filtering mechanism. In the Frenet coordinate system, road boundaries, ego states, and static and dynamic obstacles are represented uniformly to construct an S–L fused risk field and an S–T spatiotemporal interaction graph, enabling the filtering of temporally irrelevant conflict regions based on TTA relationships. At the path-planning layer, risk-guided adaptive sampling is integrated with dynamic programming and quadratic programming to improve search efficiency and trajectory quality. At the speed-planning layer, spatiotemporal coordination is achieved through non-uniform discretization, safe-corridor extraction, and speed-profile optimization. Simulation results show that the proposed method generates safe, smooth, continuous, and executable local trajectories in scenarios involving static-obstacle avoidance, adjacent-vehicle cut-ins, non-motorized road-user crossings, and mixed multi-obstacle interactions, while reducing unnecessary deceleration and detours. Ablation results further indicate that adaptive sampling reduces the number of DP search nodes by approximately 50% and the average planning time by about 30%, while maintaining a nearly unchanged minimum safety distance. These findings demonstrate that the proposed framework effectively suppresses redundant conflict regions and improves planning efficiency, solution feasibility, and motion continuity without compromising safety. Full article

As a core load-bearing component, the aero-engine rear cooling plate requires its design to simultaneously meet strength requirements and lightweight indicators. The topology optimization method considering stress constraints is the core technical path to achieve this goal, but it suffers from insufficient control precision in key areas, easily leading to material redundancy. To address this issue, a partitioned topology optimization method based on the multi-feature K-means algorithm is proposed. First, by integrating multi-dimensional features including element stress, physical density, and spatial position, an innovative multi-feature K-means algorithm is employed to realize dynamic adaptive partitioning during the optimization process. Secondly, combined with the p-norm method for partitioned stress aggregation, a precise prediction and control method for partitioned stress is adopted to refine stress constraints. Thirdly, a topology optimization model of the rear cooling plate with multi-feature partitioned stress constraints is constructed, and the adjoint method is used to solve the stress sensitivities under centrifugal loads. Finally, the effectiveness of the proposed method is verified using the rear cooling plate model. The rear cooling plate is discretized with 0.5 mm 2D axisymmetric finite elements, the filter radius is 4 mm, and the Method of Moving Asymptotes (MMA) is employed for the solution. The mass fraction of the finally optimized rear cooling plate structure is 0.157, which is 13.7% lower than that obtained by the global stress constraint method and 11.3% lower than that obtained by the topology optimization method considering both the geometric partitioned stress constraints and global stress constraints. The proposed method provides a new approach for the lightweight design of the aero-engine rear cooling plate. Full article

Aiming to clarify the spatial distribution characteristics of soil microbial assemblages and the environmental factors shaping them across a narrow altitudinal transect, this investigation concentrated on the surface soil layer within naturally occurring mixed forests of Picea crassifolia and Betula platyphylla, situated in the elevation band from 2400 to 2800 m along the southern flank of the Qilian Mountains. Leveraging the Illumina NextSeq 2000 high-throughput sequencing platform, integrated with α- and β-diversity analyses and redundancy analysis (RDA), we systematically characterized the composition and diversity traits of soil bacterial and fungal communities, as well as their associations with environmental factors. Notably, the bacterial communities were dominated by Pseudomonadota, Actinomycetota, and Acidobacteria with the abundance of Pseudomonadota decreasing with increasing altitude and that of Acidobacteria increasing with increasing altitude. Furthermore, Ascomycota and Basidiomycota were the dominant phyla in the fungal community. In contrast, bacterial α-diversity—as estimated by the Ace index—showed no significant variation across altitudes. Yet, the fungal alpha diversity metrics—namely Ace and Chao1—were markedly elevated at the 2800 m elevation relative to those observed at both intermediate and lower-altitude locations. Importantly, fungal diversity and community composition showed stronger altitudinal differentiation than bacterial communities in this dataset. Moreover, soil pH, total phosphorus, organic carbon, litter C:N:P stoichiometric ratios, and microbial biomass C:N:P stoichiometric ratios were strongly associated with soil microbial community variation along the altitude gradient, suggesting that they may act as important environmental filters. In conclusion, altitude-driven variations in litter characteristics and soil physicochemical properties jointly shape the assembly processes and spatial distribution patterns of soil microbial communities in this region. Full article