Search Results (93)

Image retrieval often begins with candidate generation, where a small subset of database images is selected for subsequent reranking or inspection. Standard multi-table p-stable locality-sensitive hashing (LSH) is a classical solution for this stage, but its isotropically sampled projection vectors do not explicitly exploit the principal structure of deep feature distributions. To address this limitation, this paper proposes a PCA-guided candidate generation method using the multi-table bucketization and query architecture of p-stable LSH. Unlike conventional PCA-based feature extraction, PCA is not used here to transform the image features themselves, but to bias the sampling distribution of projection vectors during indexing. Experiments on CALTECH101, CIFAR-10, and Tiny-ImageNet, performed using VGG19, ConvNeXt-Tiny, and ViT-B/16 features, show that the proposed method generally improves on the standard p-stable baseline across different settings. Additional comparisons with HNSW and IVF further position the method relative to modern ANN baselines. The results indicate that PCA-guided projection is an effective lightweight enhancement for systems based on the multi-table p-stable LSH architecture, improving candidate coverage and robustness while remaining compatible with the original online query process. Full article

Relational databases power high-stakes decisions in lending, healthcare, and justice, yet SQL lacks native constructs for causal and counterfactual reasoning. Prior SQL-based causal systems address parts of this gap but do not unify treatment-effect estimation with counterfactual generation in a single, composable SQL surface. We present a system that extends the SQL dialect with two declarative operators: EXPLAIN_CAUSALLY_WHY ($\psi$) for estimating average and conditional treatment effects via meta-learners, and EXPLAIN_COUNTERFACTUAL ($\phi$) for generating diverse, constraint-respecting alternatives via a hybrid KD-tree/LSH pipeline. Both operators consume standard SQL relations (joins, filters, projections) and return table-valued results with optional diagnostics, confidence intervals, and feasibility metrics. We formalize the operators in relational algebra, and describe our prototype system called PsiQL. On four evaluation datasets, PsiQL recovers a protective TWINS treatment effect, returns a non-significant COMPAS point ATE with imbalance diagnostics, flags HMDA covariate imbalance via built-in SMD checks, and generates constraint-respecting counterfactuals; a synthetic Census run serves as a balanced pipeline proof-of-concept alongside a real ACS diagnostic under severe imbalance. Full article

This paper presents high-performance Advanced Vector Extensions 512 (AVX-512) implementations of two Korean standard cryptographic algorithms: the Lightweight Secure Hash (LSH) function and Lightweight Encryption Algorithm–Galois/Counter Mode (LEA-GCM) authenticated encryption. For LSH, we apply three optimization strategies: single-message processing using AVX-512 512-bit vector registers, dual-message parallel processing through register interleaving, and multi-core parallelization using a dynamic queue-based pthread thread pool. For LEA-GCM, we propose an end-to-end optimization that replaces scalar counter-mode (CTR) encryption with 16-block AVX-512 parallel processing and Streaming SIMD Extensions(SSE)-based Galois Hash (GHASH) authentication with VPCLMULQDQ-based four-block parallel processing. Performance evaluation on an Intel Core i7-1165G7 (Tiger Lake) processor shows that LSH-256 achieves an average of 1.16× throughput improvement and LSH-512 achieves an average of 1.61× improvement over the Korea Internet and Security Agency (KISA) AVX2 reference implementation. Dual-message interleaving achieves an average superlinear speedup of 2.28×, driven by instruction-level parallelism (ILP), and the thread pool delivers speedups of 3.73× to 5.13× across eight logical cores (four physical cores with hyperthreading). The optimized LEA-GCM implementation achieves a 3.26× throughput improvement over the KISA SSE-based reference and a 12.1× improvement over the pure software implementation for 4096-byte inputs, with correctness verified against KISA official test vectors. Full article

Accurate cultivar identification of commercial dried goji berries is essential for raw material sorting, batch consistency assessment, and quality control during processing and distribution. Conventional approaches based on manual judgment or physicochemical analysis are often subjective, labor-intensive, time-consuming, and costly, making them unsuitable for rapid commercial sorting and quality inspection. To develop a rapid, low-cost, and nondestructive method for dried goji berry cultivar identification, this study proposes a visual recognition framework that integrates RGB imaging with lightweight deep learning. A dataset comprising 25,899 RGB images from five cultivars of commercial dried goji berry samples, namely Ningqi No. 7, Linqi No. 5, Ningqi No. 1, Keqi 6082, and Jingqi No. 1, was constructed. Given the pronounced surface shrinkage, complex texture, and subtle inter-cultivar appearance differences of dried goji berries, an image quality enhancement method was designed to strengthen the representation of color gradation, textural details, and edge information. For model development, CoAtNet was selected as the baseline network and redesigned for lightweight deployment. By integrating an improved feature extraction module and an information-preserving downsampling module, the proposed LSH-CoAtNet model enhances fine-grained feature representation while reducing computational cost. On the quality-enhanced image dataset, the proposed method achieved an accuracy of 98.80%, a precision of 98.81%, a recall of 98.80%, and an F1-score of 98.80%. The model contained only 6.41 M parameters and required 1.60 GFLOPs, outperforming the baseline model in both classification performance and computational efficiency. Ablation experiments and five-fold cross-validation further confirmed the effectiveness of the image quality enhancement strategy, the contribution of each improved module, and the stability of the model. Overall, the proposed method, which combines RGB image quality enhancement with LSH-CoAtNet, provides a low-cost, nondestructive, and efficient technical solution for rapid cultivar identification, raw material sorting, batch consistency assessment, and quality control of commercial dried goji berries during processing and distribution. It may also serve as a reference for intelligent classification and quality inspection of other specialty dried horticultural products. Full article

Sensor-based human activity recognition (HAR) plays a fundamental role in healthcare monitoring, sports analytics, and ambient-assisted living. Although deep learning has substantially advanced HAR performance, two practical issues still limit its real-world deployment: (i) the distribution shift caused by changes in users or sensor placements can degrade generalization, and (ii) the quadratic $\mathcal{O}\left(L^2\right)$ complexity of standard self-attention hinders efficient long-sequence modeling on resource-constrained wearable devices. To address these issues, we propose DSHformer, which is an accuracy-oriented HAR framework that combines compact channel–temporal encoding with locality-sensitive hashing (LSH)-based attention. Specifically, DSHformer (i) employs a low-parameter patch-based graph-attention encoder to jointly model latent relationships among sensor channel–temporal dynamics; (ii) introduces a trainable prototype pool together with a multi-layer decomposition network to improve intra-class compactness and inter-class separability via prototype alignment; and (iii) introduces a decomposition-stable LSH-based attention mechanism tailored for HAR, whose core design couples prototype-guided feature decomposition with locality-sensitive hashing to ensure that semantically related tokens remain consistently grouped in the same hash bucket even after decomposition-induced attenuation. The mechanism thereby operates at $\mathcal{O}(LlogL)$ attention complexity on longer sensor sequences. Extensive experiments on five public benchmarks (WISDM, UCI-HAR, PAMAP2, Opportunity, and UniMiB-SHAR) show that DSHformer achieves accuracies of 98.6%, 93.7%, 98.4%, 88.5%, and 96.6%, respectively, achieving competitive or superior performance compared with both Transformer variants and HAR-specific baselines under the adopted benchmark protocols. Ablation studies further confirm the complementary contribution of each component. Full article

Locality-sensitive hashing (LSH) has been widely used for c-approximate nearest neighbor search (c-ANNS) in high-dimensional spaces. However, its retrieval performance degrades as the dataset size grows. To address this limitation, we propose a data-dependent hashing framework called data-aware locality-sensitive hashing (DASH). DASH exploits the quantization properties of product quantization (PQ) to learn a data-aware residual prior, enabling adaptive, data-sensitive LSH. By converting exact Euclidean distance computations into efficient table lookup operations, DASH reduces computational costs and enhances retrieval efficiency. Based on the residual prior, DASH provides a theoretical performance guarantee comparable to that of standard LSH. Extensive experiments on multiple benchmark datasets demonstrate that DASH consistently achieves superior search accuracy and efficiency, yielding up to 40× speedups over various state-of-the-art baselines. Full article

Post-disaster Search and Rescue (SAR) missions demand rapid coordination of Heterogeneous Unmanned Aerial Vehicle (UAV) fleets under stringent payload and flight range limitations. Traditional heuristic solvers struggle to solve the Large-Scale Heterogeneous Team Orienteering Problem (LSH-TOP) within operational time limits due to the coupled complexity of task allocation and route planning. A Hierarchical Deep Reinforcement Learning framework decomposes this high-dimensional combinatorial problem into tractable sub-problems. An upper-level policy, guided by Monte Carlo Tree Search (MCTS), partitions the global target set to balance fleet workload distribution, whereas a lower-level Transformer-based model constructs near-optimal trajectories for individual agents. A Curriculum-Integrated Alternating Cooperative Training (C-ACT) protocol resolves the convergence difficulties associated with sparse feasible solutions in constrained environments. This protocol incorporates a dynamic constraint annealing strategy and a virtual agent buffer to progressively shape the solution space from relaxed to strictly constrained formulations. Experiments conducted on real-world geographic data demonstrate the proposed approach consistently outperforms all baselines across scales of 80 to 300 targets, improving over the strongest competitor by 0.63–8.51% and over conventional heuristics by up to 53.27% in objective value. Results indicate a task completion rate of 27.5% at the 300-target scale (versus 25.1% for the strongest baseline MCTS + OR) and balanced workload distribution, validating framework adaptability to complex emergency response scenarios. Full article

Wafer-level testing of Photonic Integrated Circuits (PICs) represents a critical throughput bottleneck in silicon photonics manufacturing, particularly as co-packaged optics demand testing of thousands of optical I/O per wafer. This work introduces optimized alignment algorithms for the Technoprobe Eclipse Dynamic probe card system, which integrates electrical probes and a piezoelectrically actuated fiber array unit within a single probe head, eliminating external positioning equipment. We systematically evaluate seven alignment algorithms: Reference Coarse Scan, Reference Coarse+Fine Scan, Cross Scan, Local and Global Bayesian Optimization, Variable and Fixed Gradient Ascent. The evaluation is made across 72 simulated test cases derived from eight experimental datasets through systematic spatial windowing, combined with experimental validation. Performance is assessed under four operating regimes—high-speed (HS) and low-speed (LS) operation, each with or without hysteresis compensation (H/NH). Experimental validation across eight die positions confirms 100% success rate for both Local Bayesian (98.24% accuracy in 99.87 arbitrary units (a.u.)) and Fixed Gradient (99.18% accuracy in 154.01 a.u.) baseline algorithms. Comprehensive simulation results with improved algorithms across all four scenarios reveal distinct performance characteristics. Fixed Gradient achieves the highest reliability (95.8%) with 99.4% average accuracy across all operating conditions. Variable Gradient provides the fastest alignment (1.18 a.u. in HS-NH) with 90.3% reliability. Local Bayesian demonstrates 94.4% reliability with intermediate performance. Global Bayesian Optimization achieves the best sample efficiency (average 24 steps) but exhibits scenario-dependent reliability ranging from 88.9% (HS-H, LS-H) to 93.1% (LS-NH). For the ideal production scenario, high speed with effective hysteresis compensation (HS-NH), Fixed Gradient emerges as the optimal choice, delivering 95.8% reliability with 1.44 a.u. alignment time, resulting in the best success rate while being nearly as fast as the fastest method. Variable Gradient achieves the absolute fastest alignment (1.18 a.u.) but with 5.5% lower reliability (90.3%), making it suitable only for applications tolerating higher failure rates. Under realistic production conditions with uncompensated hysteresis (HS-H), Fixed Gradient maintains its advantage (95.8% reliability, 3.32 a.u.), while Global Bayesian degrades significantly (88.9% reliability, 4.29 a.u.). Statistical analysis using data profiles validates these methods for high-volume PIC manufacturing, with the Eclipse Dynamic system demonstrating per-die optical alignments in sub-second timescales using open-loop control hardware. Full article

Real-time and accurate crop row extraction is a fundamental requirement for vision-based perception in autonomous agricultural machinery. In maize fields, however, row detection is easily affected by variable illumination, leaf occlusion, weed interference, and uneven soil backgrounds, which can reduce the reliability of both GNSS- and image-based navigation methods. To address these challenges, this study proposes a plant-oriented crop row perception framework that reconstructs row structures from individual maize plant detections. A lightweight detection model, named Maize–YOLO, was developed based on YOLOv11n for maize seedling detection. Three key improvements were introduced to enhance the balance between accuracy and efficiency. First, the C3k2_Faster_CGLU module replaces the original C3k2 block to reduce redundant convolutional computation while improving selective feature representation through convolutional gated linear units, thereby enhancing robustness under complex field backgrounds. Second, a lightweight shared detection head, Detect_LSH, was designed to share convolutional parameters across multi-scale feature maps and adaptively adjust feature amplitudes, reducing detection-head redundancy while maintaining multi-scale prediction capability. Third, a Layer-Adaptive Magnitude-Based Pruning strategy was applied to remove low-contribution channels and further improve computational efficiency for CPU-based deployment. Experimental results on field-collected maize seedling images showed that Maize–YOLO achieved an mAP@0.5 of 97.6%, reduced GFLOPs by 61.9%, and maintained a CPU inference speed of 84.4 FPS. After plant detection, row centerlines were estimated using an IPM–DBSCAN–LSM pipeline, which transformed detected plant centers into a quasi-top-view space, clustered them into crop rows, and fitted continuous centerlines. The extracted crop rows reached a positional accuracy of 98.6%, with a mean angular deviation of 0.44°. These results demonstrate that the proposed method can provide accurate, lightweight, and real-time crop row perception for autonomous agricultural navigation and precision field operations. Full article

The potato tuber is a metamorphic organ formed by the expansion of the underground stolon tip. It is an economically important organ and an excellent material for studying the occurrence and development of modified plant organs. However, genetic studies have lagged due to the potato’s complex genetic background. In this study, we used stolons and tubers of the potato ‘Qingshu 9’ at different stages of the tuberization process as samples for transcriptome sequencing and systematically analyzed the transcriptome characteristics of tuberization. Through RT-qPCR analysis, 16 candidate genes related to tuberization were identified. Overexpression verification was performed on one candidate gene, StLSH10, and the results indicated that it might be involved in regulating tuberization. This research provides a theoretical basis for elucidating the molecular mechanism of tuberization and offers a new target to improve potato yield and quality through molecular breeding strategies. Full article

This study systematically investigates the laser surface hardening (LSH) behavior of two medium carbon steels—the low alloy 42CrMo4 and the plain carbon C45—using a 4 kW high power diode laser (HPDL). The influence of laser parameters (power: 3.0–3.8 kW; scanning speed: 10–16 mm/s), post-laser quenching medium (oil vs. air), and, critically, the initial material condition (normalized “raw” vs. quenched and tempered “Q&T”) on the case hardening depth (CHD) was evaluated. Hardness profiles defined the CHD at a threshold of 392 HV1, and microstructural analysis was conducted via optical microscopy. The results demonstrate that prior conventional Q&T heat treatment of 42CrMo4 enhances the subsequent laser-hardened depth by approximately 27% compared to laser treatment of the normalized material under identical parameters, providing a quantitative basis for process optimization. For Q&T 42CrMo4, the quenching medium had an insignificant effect on CHD, with air cooling proving equally effective as oil across the tested parameter range, offering an empirically validated route for sustainable processing. In contrast, C45 exhibited a substantially lower and less parameter-sensitive CHD, constrained by its inherent low hardenability. This comparative analysis underscores that hardening depth in 42CrMo4 is linearly controllable via energy input, whereas for C45 it is hardenability-limited. This work establishes that an integrated approach combining conventional bulk heat treatment with diode laser hardening using air cooling offers a highly effective, controllable, and sustainable surface engineering route for high-performance alloy steels. Full article

The ALOG (Arabidopsis LIGHT-DEPENDENT SHORT HYPOCOTYLS 1 (LSH1) and Oryza G1) family play crucial regulatory roles in plant growth and development, spanning both vegetative and reproductive growth. This study presents a comprehensive genomic and functional analysis of the ALOG family in greater yam (Dioscorea alata). Ten non-redundant DaALOG genes were identified and classified into two classes (I and II) based on phylogenetic analysis. These classes share a common origin, and family expansion was primarily driven by segmental duplication events. Comparative genomics across 15 plant species revealed widespread, lineage-specific divergence in ALOG gene family size and composition. Expression profiling highlighted several DaALOG genes, particularly DaALOG1, DaALOG3, and DaALOG6A, with significant upregulation in tuber and bulbil tissues, suggesting a potential role in storage organ development. Co-expression network analysis, coupled with yeast one-hybrid assays, indicated that DaALOG3 likely regulates key genes involved in starch biosynthesis. Subcellular localization confirmed the nuclear predominance of DaALOG proteins. Furthermore, functional validation in Arabidopsis demonstrated that overexpression of DaALOG1 leads to pronounced developmental alterations, including irregular leaf morphology and floral organ abnormalities (such as extra stamens and petals). Collectively, our findings establish the DaALOG gene family as an important regulator in greater yam, linking specific members to both vegetative architecture and storage organ development. Full article

Laser surface hardening (LSH) is an efficient and flexible technique for improving the surface integrity of steel components used in high-load automotive applications. In this study, the surface changes occurring during laser hardening of 42CrMo4 steel were systematically investigated using a 4 kW high-power diode laser. The influence of laser power and scanning speed on surface roughness, hardness distribution, hardened layer depth, tribological behavior, and phase composition was analyzed. Surface topography was evaluated using three-dimensional laser scanning microscopy, while mechanical performance was assessed through microhardness and scratch testing. Phase transformations and residual structural changes were examined by X-ray diffraction (XRD) at different depths beneath the treated surface. The results demonstrate that laser processing parameters strongly affect surface integrity through competing mechanisms of surface melting, oxidation, and self-quenching. High laser power combined with low scanning speed produced deep hardened layers but promoted surface melting and retained austenite formation, whereas lower power and higher scanning speed yielded a stable martensitic surface with reduced roughness and a steep hardness gradient. XRD analysis confirmed that oxide formation was limited to the near-surface region, while the subsurface hardened zone consisted predominantly of martensitic/bainitic phases. An optimal processing window was identified that balances surface hardness, roughness, and microstructural stability without compromising surface integrity. These findings provide practical guidelines for optimizing diode laser hardening of 42CrMo4 steel gears in industrial automotive applications. Full article

Objectives: This study aimed to evaluate the validity, reliability, and diagnostic accuracy of the Low Sit–High Step (LS–HS) Test as an original, cost-effective, and clinically practical tool for assessing lower-extremity muscle strength and function, with a specific focus on its sensitivity in detecting early-stage sarcopenia. Methods: This cross-sectional study included 205 participants divided into four groups: probable sarcopenia, sarcopenia, and two control groups (young and middle-to-older adults). The LS–HS Test was compared across groups and against standard assessments to evaluate its efficacy in measuring lower-extremity function. Reliability was verified through Cronbach’s alpha and ICC. Multinomial logistic regression was used to determine the test’s predictive power, while ROC analysis assessed its diagnostic accuracy for sarcopenia screening. Results: The LS–HS scores were significantly higher in participants with probable sarcopenia and sarcopenia (p< 0.05). Multinomial logistic regression revealed that the LS–HS performance was a significant predictor of both probable sarcopenia and sarcopenia (p < 0.001). The test demonstrated excellent internal consistency (Cronbach’s α = 0.938) and very high inter-rater and test–retest reliability (ICC = 0.998). ROC analysis confirmed high diagnostic accuracy in distinguishing both probable sarcopenia (AUC = 0.768) and sarcopenia (AUC = 0.704) (all p< 0.01). Conclusions: The LS–HS Test is a valid, reliable, and sensitive tool for assessing lower-extremity functional capacity. Its ability to identify early functional decline, often manifesting before significant muscle mass loss, positions it as an effective alternative to traditional assessments in routine clinical practice, particularly for the early detection and monitoring of the sarcopenia spectrum. Full article

The increasing complexity of plant disease manifestations, especially in cases of multiple simultaneous infections, poses significant challenges to sustainable agriculture. To address this issue, we introduce the Apple Leaf Mixed Disease Recognition (ALMDR) model, a novel multi-task learning approach specifically designed for identifying and quantifying mixed disease infections in apple leaves. ALMDR comprises four key modules: a Group Feature Pyramid Network (GFPN) for multi-scale feature extraction, a Multi-Label Classification Head (MLCH) for disease type prediction, a Leaf Segmentation Head (LSH), and a Lesion Segmentation Head (LeSH) for precise delineation of leaf and lesion areas. The GFPN enhances the traditional Feature Pyramid Network (FPN) through differential sampling and grouping strategies, significantly improving the capture of fine-grained disease characteristics. The MLCH enables simultaneous classification of multiple diseases on a single leaf, effectively addressing the mixed infection problem. The segmentation heads (LSH and LeSH) work in tandem to accurately isolate leaf and lesion regions, facilitating detailed analysis of disease patterns. Experimental results on the Plant Pathology 2021-FGVC8 dataset demonstrate ALMDR’s effectiveness, outperforming state-of-the-art methods across multiple tasks. Our model achieves high performance in multi-label classification (F1-score of 93.74%), detection and segmentation (mean Average Precision (mAP) of 51.32% and 45.50%, respectively), and disease severity estimation ($R^2$ = 0.9757). Additionally, the model maintains this accuracy while processing 6.25 frames per second, balancing performance with computational efficiency. ALMDR demonstrates potential for real-time disease management in apple orchards, with possible applications extending to other crops. Full article