Search Results (1,936)

Drought stress is a major environmental constraint that significantly reduces wheat productivity worldwide. In this study, seventeen wheat genotypes were evaluated under well-watered and drought-stressed conditions across two consecutive years (2023–2024) in a controlled greenhouse experiment. Twenty morphological and agronomic traits were recorded, and their responses to prolonged water limitation were assessed using multivariate statistical methods, including three-way ANOVA, principal component analysis (PCA), and cluster analysis. Drought stress significantly decreased all traits except the harvest index (HI), with the most severe reductions observed in traits related to secondary spikes (e.g., grain weight reduced by 95%). The ANOVA results confirmed significant genotype × treatment (G × T) interactions for key agronomic traits, with the strongest effect observed for total grain weight (F = 7064.30, p < 0.001). A PCA reduced the 20 original variables to five principal components, explaining 87.2% of the total variance. These components reflected distinct trait groups associated with productivity, spike architecture, and development in phenology. Cluster analysis based on PCA scores grouped genotypes into three clusters with contrasting drought response profiles. A yield-based evaluation confirmed the cluster structure, distinguishing genotypes with a stable performance (average yield loss ~58%) from highly sensitive ones (~70% loss). Overall, the findings demonstrate that drought tolerance in wheat is governed by complex trait interactions. Integrating a trait-based multivariate analysis with a yield stability assessment enables the identification of genotypes with superior adaptation to water-limited environments, providing an excellent genotype background for future breeding efforts. Full article

Nitrogen (N) fertilizer management in winter wheat production faces challenges from volatilization losses and sub-optimal application strategies. This is particularly problematic in the Southern Great Plains, where environmental conditions during top-dressing periods favor N losses. This study evaluated the effects of a fertilizer placement method, enhanced-efficiency fertilizers, and application timing on grain yield and protein concentration (GPC) across six site-years in Oklahoma (2016–2018). Treatments included broadcast applications of untreated urea and SuperU^® (urease/nitrification inhibitor-treated urea). These were compared with subsurface placement using single-disc and double-disc drilling systems, applied at 67 kg N ha⁻¹ during January, February, or March. Subsurface placement increased the grain yield by 324–391 kg ha⁻¹ compared to broadcast applications at sites with favorable soil conditions. However, responses varied significantly across environments. Enhanced-efficiency fertilizers showed limited advantages over untreated urea. Benefits were most pronounced during February applications under conditions favoring volatilization losses. Application timing effects were more consistent for GPC than for the yield. Later applications (February–March) increased GPC by 0.8–1.2% compared to January applications. Treatment efficacy was strongly influenced by soil pH, equipment performance, and post-application environmental conditions. This indicates that N management benefits are highly site-specific. These findings demonstrate that subsurface placement can improve nitrogen use efficiency (NUE) under appropriate conditions. However, success depends on matching application strategies to local soil and environmental factors rather than adopting universal recommendations. Full article

Automated surface defect detection in steel manufacturing is pivotal for ensuring product quality, yet it remains an open challenge owing to the extreme heterogeneity of defect morphologies—ranging from hairline cracks and microscopic pores to elongated scratches and shallow dents. Existing approaches, whether classical vision pipelines or recent deep-learning paradigms, struggle to simultaneously satisfy the stringent demands of industrial scenarios: high accuracy on sub-millimeter flaws, insensitivity to texture-rich backgrounds, and real-time throughput on resource-constrained hardware. Although contemporary detectors have narrowed the gap, they still exhibit pronounced sensitivity–robustness trade-offs, particularly in the presence of scale-varying defects and cluttered surfaces. To address these limitations, we introduce MBY (MBDNet-Attention-YOLO), a lightweight yet powerful framework that synergistically couples the MBDNet backbone with the YOLO detection head. Specifically, the backbone embeds three novel components: (1) HGStem, a hierarchical stem block that enriches low-level representations while suppressing redundant activations; (2) Dynamic Align Fusion (DAF), an adaptive cross-scale fusion mechanism that dynamically re-weights feature contributions according to defect saliency; and (3) C2f-DWR, a depth-wise residual variant that progressively expands receptive fields without incurring prohibitive computational costs. Building upon this enriched feature hierarchy, the neck employs our proposed MultiSEAM module—a cascaded squeeze-and-excitation attention mechanism operating at multiple granularities—to harmonize fine-grained and semantic cues, thereby amplifying weak defect signals against complex textures. Finally, we integrate the Inner-SIoU loss, which refines the geometric alignment between predicted and ground-truth boxes by jointly optimizing center distance, aspect ratio consistency, and IoU overlap, leading to faster convergence and tighter localization. Extensive experiments on two publicly available steel-defect benchmarks—NEU-DET and PVEL-AD—demonstrate the superiority of MBY. Without bells and whistles, our model achieves 85.8% mAP@0.5 on NEU-DET and 75.9% mAP@0.5 on PVEL-AD, surpassing the best-reported results by significant margins while maintaining real-time inference on an NVIDIA Jetson Xavier. Ablation studies corroborate the complementary roles of each component, underscoring MBY’s robustness across defect scales and surface conditions. These results suggest that MBY strikes an appealing balance between accuracy, efficiency, and deployability, offering a pragmatic solution for next-generation industrial quality-control systems. Full article

High nighttime temperature (HNT) due to asymmetric diurnal warming threatens wheat productivity. This study evaluated the effect of HNT on wheat phenology, physiology, and yield through field and controlled environment experiments in Central Queensland, Australia. Two wheat genotypes, Faraday and AVT#6, were assessed under three sowing dates—1 May (Early), 15 June (Mid), and 1 August (Late)—within the recommended sowing window for the region. In a parallel growth chamber study, the plants were exposed to two nighttime temperature regimes, of 15 °C (normal) and 20 °C (high), with consistent daytime conditions from booting to maturity. Late sowing resulted in shortened vegetative growth and grain filling periods and increased exposure to HNT during the reproductive phase. This resulted in elevated floret sterility, lower grain weight, and up to 40% yield loss. AVT#6 exhibited greater sensitivity to HNT despite maturing earlier. Leaf gas exchange analysis revealed increased nighttime respiration (Rn) and reduced assimilation (A), resulting in higher Rn/A ratio for late-sown crops. The results from controlled environment chambers resembled trends of the field experiment, producing lower grain yield and biomass under HNT. Cumulative nighttime hours above 20 °C correlated more strongly with yield losses than daytime heat. These findings highlight the need for HNT-tolerant genotypes and optimized sowing schedules under future climate scenarios. Full article

In this paper, a novel online detection system is designed to enhance accuracy and operational efficiency in the outbound logistics of automotive components after production. The system consists of a scanning portal system and an improved YOLOv12-based detection algorithm which captures images of automotive parts passing through the scanning portal in real time. By integrating deep learning, the system enables real-time monitoring and identification, thereby preventing misdetections and missed detections of automotive parts, in this way promoting intelligent automotive part recognition and detection. Our system introduces the A2C2f-SA module, which achieves an efficient feature attention mechanism while maintaining a lightweight design. Additionally, Dynamic Space-to-Depth (Dynamic S2D) is employed to improve convolution and replace the stride convolution and pooling layers in the baseline network, helping to mitigate the loss of fine-grained information and enhancing the network’s feature extraction capability. To improve real-time performance, a GFL-MBConv lightweight detection head is proposed. Furthermore, adaptive frequency-aware feature fusion (Adpfreqfusion) is hybridized at the end of the neck network to effectively enhance high-frequency information lost during downsampling, thereby improving the model’s detection accuracy for target objects in complex backgrounds. On-site tests demonstrate that the system achieves a comprehensive accuracy of 97.3% and an average vehicle detection time of 7.59 s, exhibiting not only high precision but also high detection efficiency. These results can make the proposed system highly valuable for applications in the automotive industry. Full article

As a typical grain base with a dense population and high-level urbanization, the North China Plain (NCP) faces a serious threat to its sustainable development due to water shortage. Surface water area (SWA) is a key indicator for continuously measuring the trends of regional water resources and assessing their current status. Therefore, a deep understanding of its changing patterns and driving forces is essential for achieving the sustainable management of water resources. In this study, we examined the interannual variability and trends of SWA in the NCP from 1990 to 2023 using annual 30 m water body maps generated from all available Landsat imagery, a robust water mapping algorithm, and the cloud computing platform Google Earth Engine (GEE). The results showed that the SWA in the NCP has significantly increased over the past three decades. The continuous emergence of artificial reservoirs and urban lakes, along with the booming aquaculture industry, are the main factors driving the growth of SWA. Consequently, the expansion of artificial water bodies resulted in a significant increase in water evaporation (0.16 km³/yr). Moreover, the proportion of water evaporation to regional evapotranspiration (ET) gradually increased (0–0.7%/yr), indicating that the contribution of water evaporation from artificial water bodies to ET is becoming increasingly prominent. Therefore, it can be concluded that the ever-expanding artificial water bodies have become a new hidden danger affecting the water security of the NCP through evaporative loss and deserve close attention. This study not only provides us with a new perspective for deeply understanding the current status of water resources security in the NCP but also provides a typical case with great reference value for the analysis of water resources changes in other similar regions. Full article

Agricultural nondestructive testing technology is pivotal in safeguarding food quality assurance, safety monitoring, and supply chain transparency. While conventional optical methods such as near-infrared spectroscopy and hyperspectral imaging demonstrate proficiency in surface composition analysis, their constrained penetration depth and environmental sensitivity limit effectiveness in dynamic agricultural inspections. This review highlights the transformative potential of microwave technologies, systematically examining their operational principles, current implementations, and developmental trajectories for agricultural quality control. Microwave technology leverages dielectric response mechanisms to overcome traditional limitations, such as low-frequency penetration for grain silo moisture testing and high-frequency multi-parameter analysis, enabling simultaneous detection of moisture gradients, density variations, and foreign contaminants. Established applications span moisture quantification in cereal grains, oilseed crops, and plant tissues, while emerging implementations address storage condition monitoring, mycotoxin detection, and adulteration screening. The high-frequency branch of the microwave–millimeter wave systems enhances analytical precision through molecular resonance effects and sub-millimeter spatial resolution, achieving trace-level contaminant identification. Current challenges focus on three areas: excessive absorption of low-frequency microwaves by high-moisture agricultural products, significant path loss of microwave high-frequency signals in complex environments, and the lack of a standardized dielectric database. In the future, it is essential to develop low-cost, highly sensitive, and portable systems based on solid-state microelectronics and metamaterials, and to utilize IoT and 6G communications to enable dynamic monitoring. This review not only consolidates the state-of-the-art but also identifies future innovation pathways, providing a roadmap for scalable deployment of next-generation agricultural NDT systems. Full article

This study systematically investigates 50 μm-diameter 631 stainless steel fine wires subjected to both sequential and simultaneous electrothermomechanical loading to simulate probe spring conditions in microelectronic test environments. Under cyclic current loading (~10⁴ A/cm²), the 50 μm 631SS wire maintained electrical integrity up to 0.30 A for 15,000 cycles. Above 0.35 A, rapid oxide growth and abnormal grain coarsening resulted in surface embrittlement and mechanical degradation. Current-assisted tensile testing revealed a transition from recovery-dominated behavior at ≤0.20 A to significant thermal softening and ductility loss at ≥0.25 A, corresponding to a threshold temperature of approximately 200 °C. These results establish the endurance limit of 631 stainless steel wire under coupled thermal–mechanical–electrical stress and clarify the roles of Joule heating, oxidation, and microstructural evolution in electrical fatigue resistance. A degradation map is proposed to inform design margins and operational constraints for fatigue-tolerant, electrically stable interconnects in high-reliability probe spring applications. Full article

Multimodal sentiment analysis (MSA) faces key challenges such as incomplete modality inputs, long-range temporal dependencies, and suboptimal fusion strategies. To address these, we propose MGMR-Net, a Mamba-guided multimodal reconstruction and fusion network that integrates modality-aware reconstruction with text-centric fusion within an efficient state-space modeling framework. MGMR-Net consists of two core components: the Mamba-collaborative fusion module, which utilizes a two-stage selective state-space mechanism for fine-grained cross-modal alignment and hierarchical temporal integration, and the Mamba-enhanced reconstruction module, which employs continuous-time recurrence and dynamic gating to accurately recover corrupted or missing modality features. The entire network is jointly optimized via a unified multi-task loss, enabling simultaneous learning of discriminative features for sentiment prediction and reconstructive features for modality recovery. Extensive experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS datasets demonstrate that MGMR-Net consistently outperforms several baseline methods under both complete and missing modality settings, achieving superior accuracy, robustness, and generalization. Full article

Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm for vision tasks, showing great promise due to their computational efficiency and robust capacity to model global dependencies. However, most existing learning-based dehazing methods lack physical interpretability, leading to weak generalization. Furthermore, they typically rely on spatial features while neglecting crucial frequency domain information, resulting in incomplete feature representation. To address these challenges, we propose ScaleViM-PDD, a novel network that enhances an SSM backbone with two key innovations: a Multi-scale EfficientViM with Physical Decoupling (ScaleViM-P) module and a Dual-Domain Fusion (DD Fusion) module. The ScaleViM-P module synergistically integrates a Physical Decoupling block within a Multi-scale EfficientViM architecture. This design enables the network to mitigate haze interference in a physically grounded manner at each representational scale while simultaneously capturing global contextual information to adaptively handle complex haze distributions. To further address detail loss, the DD Fusion module replaces conventional skip connections by incorporating a novel Frequency Domain Module (FDM) alongside channel and position attention. This allows for a more effective fusion of spatial and frequency features, significantly improving the recovery of fine-grained details, including color and texture information. Extensive experiments on nine publicly available remote sensing datasets demonstrate that ScaleViM-PDD consistently surpasses state-of-the-art baselines in both qualitative and quantitative evaluations, highlighting its strong generalization ability. Full article

To address freshwater scarcity and the underutilization of low-saline water in the North China Plain, a field study was conducted to evaluate the effects of alternating sprinkler irrigation using saline and fresh water on soil water–salt dynamics and corn growth. Two salinity levels (3 and 5 g·L⁻¹, representing S1 and S2, respectively) and three irrigation strategies—saline–fresh–saline–fresh (F1), saline–fresh (F2), and mixed saline–fresh (F3)—were tested, resulting in six treatments: S1F1, S1F2, S1F3, S2F1, S2F2, and S2F3. S1F1 significantly improved soil water retention at a 30–50 cm depth and reduced surface electrical conductivity (EC) and Na⁺ concentration (p < 0.05). S1F1 also promoted more uniform Mg²⁺ distribution and limited Ca²⁺ loss. Under high salinity (5 g·L⁻¹), surface salt accumulation and ion concentration (Na⁺, Mg²⁺, and Ca²⁺) increased, particularly in S2F3. Corn growth under alternating irrigation (F1/F2) outperformed the mixed mode (F3), with S1F1 achieving the highest plant height, leaf area, grain number, and 100-grain weight. The S1F1 yield surpassed others by 0.4–3.0% and maintained a better ion balance. These results suggest that alternating irrigation with low-salinity water (S1F1) effectively regulates root-zone salinity and improves crop productivity, offering a practical strategy for the sustainable use of low-saline water resources. Full article

Video temporal grounding (VTG) aims to localize a semantically relevant temporal segment within an untrimmed video based on a natural language query. The task continues to face challenges arising from cross-modal semantic misalignment, which is largely attributed to redundant visual content in sensor-acquired video streams, linguistic ambiguity, and discrepancies in modality-specific representations. Most existing approaches rely on intra-modal feature modeling, processing video and text independently throughout the representation learning stage. However, this isolation undermines semantic alignment by neglecting the potential of cross-modal interactions. In practice, a natural language query typically corresponds to spatiotemporal content in video signals collected through camera-based sensing systems, encompassing a particular sequence of frames and its associated salient subregions. We propose a text-guided visual representation optimization framework tailored to enhance semantic interpretation over video signals captured by visual sensors. This framework leverages textual information to focus on spatiotemporal video content, thereby narrowing the cross-modal gap. Built upon the unified cross-modal embedding space provided by CLIP, our model leverages video data from sensing devices to structure representations and introduces two dedicated modules to semantically refine visual representations across spatial and temporal dimensions. First, we design a Spatial Visual Representation Optimization (SVRO) module to learn spatial information within intra-frames. It selects salient patches related to the text, capturing more fine-grained visual details. Second, we introduce a Temporal Visual Representation Optimization (TVRO) module to learn temporal relations from inter-frames. Temporal triplet loss is employed in TVRO to enhance attention on text-relevant frames and capture clip semantics. Additionally, a self-supervised contrastive loss is introduced at the clip–text level to improve inter-clip discrimination by maximizing semantic variance during training. Experiments on Charades-STA, ActivityNet Captions, and TACoS, widely used benchmark datasets, demonstrate that our method outperforms state-of-the-art methods across multiple metrics. Full article

Bird-dropping fouling and hotspot anomalies remain the most prevalent and detrimental defects in utility-scale photovoltaic (PV) plants; their co-occurrence on a single module markedly curbs energy yield and accelerates irreversible cell degradation. However, markedly disparate visual–thermal signatures of the two phenomena impede high-fidelity concurrent detection in existing robotic inspection systems, while stringent onboard compute budgets also preclude the adoption of bulky detectors. To resolve this accuracy–efficiency trade-off for dual-defect detection, we present YOLOv8-SG, a lightweight yet powerful framework engineered for mobile PV inspectors. First, a rigorously curated multi-modal dataset—RGB for stains and long-wave infrared for hotspots—is assembled to enforce robust cross-domain representation learning. Second, the HSV color space is leveraged to disentangle chromatic and luminance cues, thereby stabilizing appearance variations across sensors. Third, a single-head self-attention (SHSA) block is embedded in the backbone to harvest long-range dependencies at negligible parameter cost, while a global context (GC) module is grafted onto the detection head to amplify fine-grained semantic cues. Finally, an auxiliary bounding box refinement term is appended to the loss to hasten convergence and tighten localization. Extensive field experiments demonstrate that YOLOv8-SG attains 86.8% mAP@0.5, surpassing the vanilla YOLOv8 by 2.7 pp while trimming 12.6% of parameters (18.8 MB). Grad-CAM saliency maps corroborate that the model’s attention consistently coincides with defect regions, underscoring its interpretability. The proposed method, therefore, furnishes PV operators with a practical low-latency solution for concurrent bird-dropping and hotspot surveillance. Full article

Investigating the factors influencing rice grain yield (GY) is critical for optimizing nitrogen (N) management and enhancing resource use efficiency in rice cultivation. However, few studies have comprehensively investigated the factors affecting rice GY, considering an entire influence chain encompassing rice N uptake, growth indicators, and GY components. In this study, field experiment with six different N fertilizer rates (0, 60, 120, 180, 225, and 300 kg N ha⁻¹, i.e., N0, N60, N120, N180, N225, and N300) was conducted in the Jianghan Plain in the Middle Reaches of the Yangtze River, China, to comprehensively elucidate the factors influencing rice GY from aspects of rice N uptake, growth indicators, and GY components and determine the optimal N fertilizer rate. The results showed that rice GY and N uptake initially increased and then either stabilized or declined with higher N fertilizer rate, while apparent N loss escalated with increased N fertilizer rate. The application of N fertilizer significantly promoted the increase in straw N uptake, which was significantly positively correlated with growth indicators (p < 0.05). Among all GY components, panicle number per hill was the most significant positive factor influencing rice GY, and it was significantly positively correlated with all rice growth indicators (p < 0.05). In addition, N180 was the optimal N fertilizer rate, ensuring more than 95% of maximum GY and reducing N loss by 74% and 39% compared to N300, respectively. Meanwhile, the average N balance for N180 remained below 60 kg N ha⁻¹. In conclusion, optimizing the N fertilizer application in paddy fields can effectively maintain stable rice GY and minimize environmental pollution. Full article

Deep learning (DL), as a cutting-edge technology in artificial intelligence, has significantly impacted fields such as computer vision and natural language processing. Loss function determines the convergence speed and accuracy of the DL model and has a crucial impact on algorithm quality and model performance. However, most of the existing studies focus on the improvement of specific problems of loss function, which lack a systematic summary and comparison, especially in computer vision and natural language processing tasks. Therefore, this paper reclassifies and summarizes the loss functions in DL and proposes a new category of metric loss. Furthermore, this paper conducts a fine-grained division of regression loss, classification loss, and metric loss, elaborating on the existing problems and improvements. Finally, the new trend of compound loss and generative loss is anticipated. The proposed paper provides a new perspective for loss function division and a systematic reference for researchers in the DL field. Full article