Microorganisms can colonize the surface of microplastics(MPs)to form a distinctive microbiome,known as a“plastisphere”which is regarded as an anthropogenic niche for microbial growth.However,bacterial community asse...Microorganisms can colonize the surface of microplastics(MPs)to form a distinctive microbiome,known as a“plastisphere”which is regarded as an anthropogenic niche for microbial growth.However,bacterial community assembly in virgin and aging MP plastispheres across different habitats is poorly understood.This study aims to assess the variations in bacterial community assembly across different niches and habitats with an in situ ex-periment,in which constructed forest wetland(FW),natural lake wetland(LW),and lotus pond wetland(LP)were habitats,and plastispheres of virgin and aging low-density polyethylene(LDPE)MPs,as well as surround-ing wetland soils were niches.Significant niche-related differences in bacterial communities were observed,with lower diversity and enrichment of potential plastic-degrading bacteria in the plastisphere than in the soil bacterial communities.Furthermore,habitat-related differences exerted a more pronounced influence on the beta-diversity patterns of the bacterial communities.The linear regression analyses indicated that the local species pool con-tributed more to bacterial community assembly in the LW wetland,whereas the relative abundance of species was the primary factor in the LP wetland.The null model analysis indicated that plastisphere bacterial communi-ties were predominantly driven by the stochastic process,with a more deterministic assembly observed in the LP wetland and soil bacterial communities.Additionally,the primary ecological process shaping plastisphere com-munities shifted from drift in the virgin LDPE to homogenising dispersal in the aging LDPE.This study provides new insights into the fate and ecological impacts of MPs in wetlands,thereby facilitating the effective regulations of plastic pollution.展开更多
Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approach...Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.展开更多
Video emotion recognition is widely used due to its alignment with the temporal characteristics of human emotional expression,but existingmodels have significant shortcomings.On the one hand,Transformermultihead self-...Video emotion recognition is widely used due to its alignment with the temporal characteristics of human emotional expression,but existingmodels have significant shortcomings.On the one hand,Transformermultihead self-attention modeling of global temporal dependency has problems of high computational overhead and feature similarity.On the other hand,fixed-size convolution kernels are often used,which have weak perception ability for emotional regions of different scales.Therefore,this paper proposes a video emotion recognition model that combines multi-scale region-aware convolution with temporal interactive sampling.In terms of space,multi-branch large-kernel stripe convolution is used to perceive emotional region features at different scales,and attention weights are generated for each scale feature.In terms of time,multi-layer odd-even down-sampling is performed on the time series,and oddeven sub-sequence interaction is performed to solve the problem of feature similarity,while reducing computational costs due to the linear relationship between sampling and convolution overhead.This paper was tested on CMU-MOSI,CMU-MOSEI,and Hume Reaction.The Acc-2 reached 83.4%,85.2%,and 81.2%,respectively.The experimental results show that the model can significantly improve the accuracy of emotion recognition.展开更多
In a multiple voltage source converter(VSC)system,the nonlinear characteristics of phase-locked loops(PLLs)and their interactions have a significant influence on the synchronization stability of converters.In this pap...In a multiple voltage source converter(VSC)system,the nonlinear characteristics of phase-locked loops(PLLs)and their interactions have a significant influence on the synchronization stability of converters.In this paper,these influences are investigated from the perspective of the time domain.First,a novel time-domain model of the multi-VSC system is obtained by using a multi-scale method.On this basis,a stability criterion is proposed to assess the synchronization stability of the system.Then,the accuracy of the time-domain model and its stability criterion in various conditions are discussed.Moreover,the negative impact of the interaction on the system is quantified.Finally,the above theoretical analysis is also verified in the controller hardware-in-the-loop(CHIL)experiments.展开更多
Camouflaged Object Detection(COD)aims to identify objects that share highly similar patterns—such as texture,intensity,and color—with their surrounding environment.Due to their intrinsic resemblance to the backgroun...Camouflaged Object Detection(COD)aims to identify objects that share highly similar patterns—such as texture,intensity,and color—with their surrounding environment.Due to their intrinsic resemblance to the background,camouflaged objects often exhibit vague boundaries and varying scales,making it challenging to accurately locate targets and delineate their indistinct edges.To address this,we propose a novel camouflaged object detection network called Edge-Guided and Multi-scale Fusion Network(EGMFNet),which leverages edge-guided multi-scale integration for enhanced performance.The model incorporates two innovative components:a Multi-scale Fusion Module(MSFM)and an Edge-Guided Attention Module(EGA).These designs exploit multi-scale features to uncover subtle cues between candidate objects and the background while emphasizing camouflaged object boundaries.Moreover,recognizing the rich contextual information in fused features,we introduce a Dual-Branch Global Context Module(DGCM)to refine features using extensive global context,thereby generatingmore informative representations.Experimental results on four benchmark datasets demonstrate that EGMFNet outperforms state-of-the-art methods across five evaluation metrics.Specifically,on COD10K,our EGMFNet-P improves F_(β)by 4.8 points and reduces mean absolute error(MAE)by 0.006 compared with ZoomNeXt;on NC4K,it achieves a 3.6-point increase in F_(β).OnCAMO and CHAMELEON,it obtains 4.5-point increases in F_(β),respectively.These consistent gains substantiate the superiority and robustness of EGMFNet.展开更多
Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectra...Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectral similarity between buildings and backgrounds,sensor variations,and insufficient computational efficiency.To address these challenges,this paper proposes a novel Multi-scale Efficient Wavelet-based Change Detection Network(MewCDNet),which integrates the advantages of Convolutional Neural Networks and Transformers,balances computational costs,and achieves high-performance building change detection.The network employs EfficientNet-B4 as the backbone for hierarchical feature extraction,integrates multi-level feature maps through a multi-scale fusion strategy,and incorporates two key modules:Cross-temporal Difference Detection(CTDD)and Cross-scale Wavelet Refinement(CSWR).CTDD adopts a dual-branch architecture that combines pixel-wise differencing with semanticaware Euclidean distance weighting to enhance the distinction between true changes and background noise.CSWR integrates Haar-based Discrete Wavelet Transform with multi-head cross-attention mechanisms,enabling cross-scale feature fusion while significantly improving edge localization and suppressing spurious changes.Extensive experiments on four benchmark datasets demonstrate MewCDNet’s superiority over comparison methods:achieving F1 scores of 91.54%on LEVIR,93.70%on WHUCD,and 64.96%on S2Looking for building change detection.Furthermore,MewCDNet exhibits optimal performance on the multi-class⋅SYSU dataset(F1:82.71%),highlighting its exceptional generalization capability.展开更多
Gold nanoclusters(AuNCs)are ultrasmall(<2 nm)aggregates of gold atoms that exhibit discrete electronic states,size-dependent photoluminescence,and exceptional biocompatibility,making them ideal candidates for thera...Gold nanoclusters(AuNCs)are ultrasmall(<2 nm)aggregates of gold atoms that exhibit discrete electronic states,size-dependent photoluminescence,and exceptional biocompatibility,making them ideal candidates for theranostic applications.Their tunable surface chemistry enables targeted delivery,while strong near-infrared emission and environmental responsiveness allow for sensitive detection and deep-tissue imaging.Recent advances have revealed that controlled assembly of AuNCs into higher-order architectures-guided by biological scaffolds such as nucleic acids,peptides,and proteins-can markedly enhance their optical and electronic properties through aggregation-induced emission(AiE)and stabilization of surface ligands.This review summarizes recent progress in the design and biomedical applications of AuNC assemblies generated using biomolecules as structure-directing scaffolds.Covalent and noncovalent interactions with biomolecules enable the formation of well-defined one-,two-,and three-dimensional structures with tunable morphologies and sizes.These assemblies display distinctive photophysical behaviors that have been exploited for biosensing,bioimaging,and therapeutic applications in both cellular and in vivo models.Compared with individual AuNCs,assembled systems offer improved uptake,prolonged circulation,and efficient clearance,while protecting labile cargos such as nucleic acids and proteins.Moreover,their ordered and defined architectures can be engineered for controlled drug release and synergistic photo-or radiotherapeutic effects.Despite these advances,fundamental understanding of how structural organization governs photophysical responses remains limited.Elucidating parameters such as intercluster spacing and loading density will be essential for optimizing performance.Overall,biologically guided AuNC assemblies represent a powerful platform for multifunctional biosensing and therapy,bridging nanoscale design with biological function.展开更多
Tomato is a major economic crop worldwide,and diseases on tomato leaves can significantly reduce both yield and quality.Traditional manual inspection is inefficient and highly subjective,making it difficult to meet th...Tomato is a major economic crop worldwide,and diseases on tomato leaves can significantly reduce both yield and quality.Traditional manual inspection is inefficient and highly subjective,making it difficult to meet the requirements of early disease identification in complex natural environments.To address this issue,this study proposes an improved YOLO11-based model,YOLO-SPDNet(Scale Sequence Fusion,Position-Channel Attention,and Dual Enhancement Network).The model integrates the SEAM(Self-Ensembling Attention Mechanism)semantic enhancement module,the MLCA(Mixed Local Channel Attention)lightweight attention mechanism,and the SPA(Scale-Position-Detail Awareness)module composed of SSFF(Scale Sequence Feature Fusion),TFE(Triple Feature Encoding),and CPAM(Channel and Position Attention Mechanism).These enhancements strengthen fine-grained lesion detection while maintaining model lightweightness.Experimental results show that YOLO-SPDNet achieves an accuracy of 91.8%,a recall of 86.5%,and an mAP@0.5 of 90.6%on the test set,with a computational complexity of 12.5 GFLOPs.Furthermore,the model reaches a real-time inference speed of 987 FPS,making it suitable for deployment on mobile agricultural terminals and online monitoring systems.Comparative analysis and ablation studies further validate the reliability and practical applicability of the proposed model in complex natural scenes.展开更多
Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posi...Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posing a significant challenge for accurate low-light object detection.Existing methods struggle to preserve discriminative features while maintaining semantic consistency between low-light and normal-light images.For this purpose,this study proposes a DL-YOLO model specially tailored for low-light detection.To mitigate target feature attenuation introduced by repeated downsampling,we design aMulti-Scale FeatureConvolution(MSF-Conv)module that captures rich,multi-level details via multi-scale feature learning,thereby reducing model complexity and computational cost.For feature fusion,we integrated the C3k2-DWRmodule by embedding the Dilation-wise Residual(DWR)mechanism into the 2-core optimized Cross Stage Partial(C3)framework,achieving efficient feature integration.In addition,we replace conventional localization losses with WIoU(Weighted Intersection over Union),which dynamically adjusts gradient gain according to sample quality,thereby improving localization robustness and precision.Experiments on the ExDark dataset demonstrate that DL-YOLO delivers strong low-light detection performance.The relevant code is published at https://github.com/cym0997/DL-YOLO.展开更多
Targeted protein degradation(TPD)is an innovative strategy for selectively eliminating pathogenic proteins,enabling precise degradation of once-undruggable targets in cancer therapy.However,current TPD molecules are o...Targeted protein degradation(TPD)is an innovative strategy for selectively eliminating pathogenic proteins,enabling precise degradation of once-undruggable targets in cancer therapy.However,current TPD molecules are often limited by poor tumor targeting and the need for high doses.To overcome these limitations,assembly/disassembly-based TPD systems have been proposed to effectively degrade proteins of interest and enhance therapeutic efficacy.Herein,we summarize the recent advances in such TPD systems and categorize the strategies employed,including nanosphere morphology of assembled TPD systems,nanofiber morphology of assembled TPD systems,carrier-mediated TPD release systems,and stimulus-induced free TPD molecule formation nanosystems.Finally,we outline future directions and identify the remaining challenges in assembly/disassembly-based TPD systems.展开更多
Ecological floating bed is an important biological remediation method for water pollution control.During the removal of excess nutrients and pollutants,changes in environmental factors affect the characteristics of mi...Ecological floating bed is an important biological remediation method for water pollution control.During the removal of excess nutrients and pollutants,changes in environmental factors affect the characteristics of microorganisms in aquatic ecosystems.To understand the influences of ecological floating beds on size-fractionated microorganisms,we investigated the community assembly and nitrogen metabolic characteristics of three size-fractionated microorganism groups in the ecological floating bed area,using 18S rDNA,16S rDNA metabarcoding,and metagenomic sequencing techniques.Firstly,we discovered substantial differences between size-fractionated groups in the diversity and compositions of both microeukaryotic and bacterial communities,as well as the influences of floating beds on specific groups.The floating beds appeared to provide more habitats for heterotrophs and symbiotes while potentially inhibiting the growth of certain phytoplankton(cyanobacteria).Secondly,we observed that microeukaryotic and bacterial communities were predominantly influenced by stochastic and deterministic processes,respectively,and they both exhibited distinct patterns across different size-fractionated groups.Notably,microeukaryotic community assembly demonstrated a greater sensitivity to ecological floating beds,as indicated by an increase in dispersal limitation processes.Finally,the nitrogen metabolism functional genes revealed that microbes associated with large-sized particles played a crucial role in dissimilatory nitrate reduction to ammonium(DNRA)and denitrification processes within the floating bed area,thereby facilitating the removal of excess nitrogen nutrients from the water.In contrast,freeliving microorganisms from small-sized groups were linked mainly to the genes involved in nitrogen assimilation and assimilatory nitrate reduction to ammonium(ANRA)processes.These findings help understand the impact of ecological floating beds on the diversity and functional characteristics of microorganism communities in different size-fractionated groups.展开更多
Semantic segmentation for mixed scenes of aerial remote sensing and road traffic is one of the key technologies for visual perception of flying cars.The State-of-the-Art(SOTA)semantic segmentation methods have made re...Semantic segmentation for mixed scenes of aerial remote sensing and road traffic is one of the key technologies for visual perception of flying cars.The State-of-the-Art(SOTA)semantic segmentation methods have made remarkable achievements in both fine-grained segmentation and real-time performance.However,when faced with the huge differences in scale and semantic categories brought about by the mixed scenes of aerial remote sensing and road traffic,they still face great challenges and there is little related research.Addressing the above issue,this paper proposes a semantic segmentation model specifically for mixed datasets of aerial remote sensing and road traffic scenes.First,a novel decoding-recoding multi-scale feature iterative refinement structure is proposed,which utilizes the re-integration and continuous enhancement of multi-scale information to effectively deal with the huge scale differences between cross-domain scenes,while using a fully convolutional structure to ensure the lightweight and real-time requirements.Second,a welldesigned cross-window attention mechanism combined with a global information integration decoding block forms an enhanced global context perception,which can effectively capture the long-range dependencies and multi-scale global context information of different scenes,thereby achieving fine-grained semantic segmentation.The proposed method is tested on a large-scale mixed dataset of aerial remote sensing and road traffic scenes.The results confirm that it can effectively deal with the problem of large-scale differences in cross-domain scenes.Its segmentation accuracy surpasses that of the SOTA methods,which meets the real-time requirements.展开更多
The Qingtongxia Irrigation District in Ningxia is an important hydrological and ecological region.To assess its ecological environment quality from 2001 to 2021 across multiple scales and identify driving factors,a mo...The Qingtongxia Irrigation District in Ningxia is an important hydrological and ecological region.To assess its ecological environment quality from 2001 to 2021 across multiple scales and identify driving factors,a modified remote sensing ecological index(MRSEI)was developed by incorporating evapotranspiration.Spatial and temporal patterns were analyzed using the coefficient of variation,spatial autocorrelation,and semi-variogram methods,while influencing factors were explored via the optimal parameter geographical detector model.The MRSEI’s first principal component loadings and rankings aligned with those of RSEI(average contribution:81.31%),effectively reflecting spatiotemporal variations.At sub-irrigation district and landscape scales,ecological quality was slightly lower than at the district level but remained stable.Moderate and good ecological grades accounted for 36.28%and 33.38%of the area,respectively,at the district scale,and the moderate grade reached 70.48%on smaller scales.Spatial heterogeneity intensified with decreasing scale,and human activity lost explanatory power below a 5 km range.Human factors mainly drove ecological differentiation at the district scale,while natural factors dominated at finer scales.The MRSEI offers a novel tool for ecological assessment in arid/semi-arid areas and supports scale-adapted ecological protection strategies.展开更多
Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)t...Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.展开更多
Advanced healthcare monitors for air pollution applications pose a significant challenge in achieving a balance between high-performance filtration and multifunctional smart integration.Electrospinning triboelectric n...Advanced healthcare monitors for air pollution applications pose a significant challenge in achieving a balance between high-performance filtration and multifunctional smart integration.Electrospinning triboelectric nanogenerators(TENG)provide a significant potential for use under such difficult circumstances.We have successfully constructed a high-performance TENG utilizing a novel multi-scale nanofiber architecture.Nylon 66(PA66)and chitosan quaternary ammonium salt(HACC)composites were prepared by electrospinning,and PA66/H multiscale nanofiber membranes composed of nanofibers(≈73 nm)and submicron-fibers(≈123 nm)were formed.PA66/H multi-scale nanofiber membrane as the positive electrode and negative electrode-spun PVDF-HFP nanofiber membrane composed of respiration-driven PVDF-HFP@PA66/H TENG.The resulting PVDF-HFP@PA66/H TENG based air filter utilizes electrostatic adsorption and physical interception mechanisms,achieving PM_(0.3)filtration efficiency over 99%with a pressure drop of only 48 Pa.Besides,PVDF-HFP@PA66/H TENG exhibits excellent stability in high-humidity environments,with filtration efficiency reduced by less than 1%.At the same time,the TENG achieves periodic contact separation through breathing drive to achieve self-power,which can ensure the long-term stability of the filtration efficiency.In addition to the air filtration function,TENG can also monitor health in real time by capturing human breathing signals without external power supply.This integrated system combines high-efficiency air filtration,self-powered operation,and health monitoring,presenting an innovative solution for air purification,smart protective equipment,and portable health monitoring.These findings highlight the potential of this technology for diverse applications,offering a promising direction for advancing multifunctional air filtration systems.展开更多
Defect detection in printed circuit boards(PCB)remains challenging due to the difficulty of identifying small-scale defects,the inefficiency of conventional approaches,and the interference from complex backgrounds.To ...Defect detection in printed circuit boards(PCB)remains challenging due to the difficulty of identifying small-scale defects,the inefficiency of conventional approaches,and the interference from complex backgrounds.To address these issues,this paper proposes SIM-Net,an enhanced detection framework derived from YOLOv11.The model integrates SPDConv to preserve fine-grained features for small object detection,introduces a novel convolutional partial attention module(C2PAM)to suppress redundant background information and highlight salient regions,and employs a multi-scale fusion network(MFN)with a multi-grain contextual module(MGCT)to strengthen contextual representation and accelerate inference.Experimental evaluations demonstrate that SIM-Net achieves 92.4%mAP,92%accuracy,and 89.4%recall with an inference speed of 75.1 FPS,outperforming existing state-of-the-art methods.These results confirm the robustness and real-time applicability of SIM-Net for PCB defect inspection.展开更多
The pore structure of shale oil reservoir significantly affects the occurrence and mobility of hydrocarbons.The potential of a new type of alkaline lake shale oil has been demonstrated,but there are few reports on the...The pore structure of shale oil reservoir significantly affects the occurrence and mobility of hydrocarbons.The potential of a new type of alkaline lake shale oil has been demonstrated,but there are few reports on the pore system of alkaline lake shale,which restricts the efficient exploration and development of shale oil.This study investigates the Fengcheng Formation shale in the Mahu sag of the Junggar Basin,employing methods such as low-temperature nitrogecn adsorption(LTNA),mercury intrusion capillary pressure(MICP),and nuclear magnetic resonance(NMR)to quantitatively characterize the multi-scale pore structure and fractal characteristics of shale,while evaluating the applicability of these methods.Based on a comprehensive analysis of material composition,different pore types,and fractal dimensions,the controlling factors for the development of different pore types and their seepage capacity are discussed.The results indicate that inorganic mineral pores are the main development in alkaline lake shale,with the pore morphology being characterized by slit-like and ink-bottle shapes.The multi-scale pore size distribution(PSD)shows that Ⅱ-micropores(10-100 nm)and mesopores(100–1000 nm)are the main contributors to the pore system.The development of Ⅱ-micropores is associated with feldspar and calcareous minerals,the development of Ⅰ-micropores(<10 nm)and mesopores is related to quartz content,while large pores are mainly found in interlayer fissures of clay minerals.The development of Ⅰ-micropores increases the roughness of pore surface and enhances the adsorption capacity of the pores,while the development of Ⅱ-micropores associated with calcareous minerals hinders pore seepage capacity.Mesopores and macropores(>1000 nm)exhibit good flowability.The high content of siliceous minerals plays a positive role in the pore system of alkaline lake shale.The shale with higher fractal dimension Dmin exhibits greater adsorption capacity,which hinders the accumulation of free-state shale oil.Different types of pore space play different roles in the occurrence of shale oil,with free-state shale oil primarily occurring in micro-fractures and inorganic mineral pores,and the pore size is exceeding 10 nm.展开更多
Transformer-based models have significantly advanced binary code similarity detection(BCSD)by leveraging their semantic encoding capabilities for efficient function matching across diverse compilation settings.Althoug...Transformer-based models have significantly advanced binary code similarity detection(BCSD)by leveraging their semantic encoding capabilities for efficient function matching across diverse compilation settings.Although adversarial examples can strategically undermine the accuracy of BCSD models and protect critical code,existing techniques predominantly depend on inserting artificial instructions,which incur high computational costs and offer limited diversity of perturbations.To address these limitations,we propose AIMA,a novel gradient-guided assembly instruction relocation method.Our method decouples the detection model into tokenization,embedding,and encoding layers to enable efficient gradient computation.Since token IDs of instructions are discrete and nondifferentiable,we compute gradients in the continuous embedding space to evaluate the influence of each token.The most critical tokens are identified by calculating the L2 norm of their embedding gradients.We then establish a mapping between instructions and their corresponding tokens to aggregate token-level importance into instructionlevel significance.To maximize adversarial impact,a sliding window algorithm selects the most influential contiguous segments for relocation,ensuring optimal perturbation with minimal length.This approach efficiently locates critical code regions without expensive search operations.The selected segments are relocated outside their original function boundaries via a jump mechanism,which preserves runtime control flow and functionality while introducing“deletion”effects in the static instruction sequence.Extensive experiments show that AIMA reduces similarity scores by up to 35.8%in state-of-the-art BCSD models.When incorporated into training data,it also enhances model robustness,achieving a 5.9%improvement in AUROC.展开更多
Fungi play crucial roles in nutrient acquisition,plant growth promotion,and the enhancement of resistance to both abiotic and biotic stresses.However,studies on the fungal communities associated with peas (Pisum sativ...Fungi play crucial roles in nutrient acquisition,plant growth promotion,and the enhancement of resistance to both abiotic and biotic stresses.However,studies on the fungal communities associated with peas (Pisum sativum L.) remain limited.In this study,we systematically investigated the ecological effects of host niches (soil,root,stem,leaf,and pod) and genotypes on the diversity and composition of fungal communities in peas using a multi-level approach that encompassed pattern recognition (β-diversity decomposition),mechanism validation (neutral community model testing),and dynamic tracking methods (migration pathway source-tracking).The results revealed that the dominant fungal phyla across niches and genotypes were Ascomycota,Basidiomycota,and Mortierellomycota,and the community structures of the soil–plant continuum were primarily determined by the pea niches rather than genotypes.β-diversity decomposition was largely attributed to species replacement rather than richness differences,indicating strong niche specificity and microbial replacement across microhabitats.Neutral model analysis revealed that stochastic processes influenced genotypeassociated communities,while deterministic processes played a dominant role in niche-based community assembly.Source-tracking analysis identified niche-to-niche fungal migration,with Erysiphe,Fusarium,Cephaliophora,Ascobolus,Alternaria,and Aspergillus as the key genera.Migration rates from exogenous to endogenous niches were low (1.3–61.5%),whereas those within exogenous (64.4–83.7%) or endogenous (73.9–96.4%) compartments were much higher,suggesting that the pea epidermis acts as a selective barrier that filters and enriches microbial communities prior to internal colonization.This study provides comprehensive insights into the mechanisms of host filtering,enrichment and microbial sourcing,which increases our understanding of the assembly rules of the pea-associated fungal microbiome.展开更多
基金supported by Shanghai Municipal Natural Science Foundation,China(No.21ZR1446800)the National Natural Science Foundation of China(No.41877425)the Fundamental Research Funds for the Central Universities(No.226-2024-00052)。
文摘Microorganisms can colonize the surface of microplastics(MPs)to form a distinctive microbiome,known as a“plastisphere”which is regarded as an anthropogenic niche for microbial growth.However,bacterial community assembly in virgin and aging MP plastispheres across different habitats is poorly understood.This study aims to assess the variations in bacterial community assembly across different niches and habitats with an in situ ex-periment,in which constructed forest wetland(FW),natural lake wetland(LW),and lotus pond wetland(LP)were habitats,and plastispheres of virgin and aging low-density polyethylene(LDPE)MPs,as well as surround-ing wetland soils were niches.Significant niche-related differences in bacterial communities were observed,with lower diversity and enrichment of potential plastic-degrading bacteria in the plastisphere than in the soil bacterial communities.Furthermore,habitat-related differences exerted a more pronounced influence on the beta-diversity patterns of the bacterial communities.The linear regression analyses indicated that the local species pool con-tributed more to bacterial community assembly in the LW wetland,whereas the relative abundance of species was the primary factor in the LP wetland.The null model analysis indicated that plastisphere bacterial communi-ties were predominantly driven by the stochastic process,with a more deterministic assembly observed in the LP wetland and soil bacterial communities.Additionally,the primary ecological process shaping plastisphere com-munities shifted from drift in the virgin LDPE to homogenising dispersal in the aging LDPE.This study provides new insights into the fate and ecological impacts of MPs in wetlands,thereby facilitating the effective regulations of plastic pollution.
基金funded by the National Natural Science Foundation of China,grant numbers 52374156 and 62476005。
文摘Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.
基金supported,in part,by the National Nature Science Foundation of China under Grant 62272236,62376128in part,by the Natural Science Foundation of Jiangsu Province under Grant BK20201136,BK20191401.
文摘Video emotion recognition is widely used due to its alignment with the temporal characteristics of human emotional expression,but existingmodels have significant shortcomings.On the one hand,Transformermultihead self-attention modeling of global temporal dependency has problems of high computational overhead and feature similarity.On the other hand,fixed-size convolution kernels are often used,which have weak perception ability for emotional regions of different scales.Therefore,this paper proposes a video emotion recognition model that combines multi-scale region-aware convolution with temporal interactive sampling.In terms of space,multi-branch large-kernel stripe convolution is used to perceive emotional region features at different scales,and attention weights are generated for each scale feature.In terms of time,multi-layer odd-even down-sampling is performed on the time series,and oddeven sub-sequence interaction is performed to solve the problem of feature similarity,while reducing computational costs due to the linear relationship between sampling and convolution overhead.This paper was tested on CMU-MOSI,CMU-MOSEI,and Hume Reaction.The Acc-2 reached 83.4%,85.2%,and 81.2%,respectively.The experimental results show that the model can significantly improve the accuracy of emotion recognition.
基金supported by the Science and Technology Project of State Grid Corporation of China(5400-202199281A-0-0-00).
文摘In a multiple voltage source converter(VSC)system,the nonlinear characteristics of phase-locked loops(PLLs)and their interactions have a significant influence on the synchronization stability of converters.In this paper,these influences are investigated from the perspective of the time domain.First,a novel time-domain model of the multi-VSC system is obtained by using a multi-scale method.On this basis,a stability criterion is proposed to assess the synchronization stability of the system.Then,the accuracy of the time-domain model and its stability criterion in various conditions are discussed.Moreover,the negative impact of the interaction on the system is quantified.Finally,the above theoretical analysis is also verified in the controller hardware-in-the-loop(CHIL)experiments.
基金financially supported byChongqingUniversity of Technology Graduate Innovation Foundation(Grant No.gzlcx20253267).
文摘Camouflaged Object Detection(COD)aims to identify objects that share highly similar patterns—such as texture,intensity,and color—with their surrounding environment.Due to their intrinsic resemblance to the background,camouflaged objects often exhibit vague boundaries and varying scales,making it challenging to accurately locate targets and delineate their indistinct edges.To address this,we propose a novel camouflaged object detection network called Edge-Guided and Multi-scale Fusion Network(EGMFNet),which leverages edge-guided multi-scale integration for enhanced performance.The model incorporates two innovative components:a Multi-scale Fusion Module(MSFM)and an Edge-Guided Attention Module(EGA).These designs exploit multi-scale features to uncover subtle cues between candidate objects and the background while emphasizing camouflaged object boundaries.Moreover,recognizing the rich contextual information in fused features,we introduce a Dual-Branch Global Context Module(DGCM)to refine features using extensive global context,thereby generatingmore informative representations.Experimental results on four benchmark datasets demonstrate that EGMFNet outperforms state-of-the-art methods across five evaluation metrics.Specifically,on COD10K,our EGMFNet-P improves F_(β)by 4.8 points and reduces mean absolute error(MAE)by 0.006 compared with ZoomNeXt;on NC4K,it achieves a 3.6-point increase in F_(β).OnCAMO and CHAMELEON,it obtains 4.5-point increases in F_(β),respectively.These consistent gains substantiate the superiority and robustness of EGMFNet.
基金supported by the Henan Province Key R&D Project under Grant 241111210400the Henan Provincial Science and Technology Research Project under Grants 252102211047,252102211062,252102211055 and 232102210069+2 种基金the Jiangsu Provincial Scheme Double Initiative Plan JSS-CBS20230474,the XJTLU RDF-21-02-008the Science and Technology Innovation Project of Zhengzhou University of Light Industry under Grant 23XNKJTD0205the Higher Education Teaching Reform Research and Practice Project of Henan Province under Grant 2024SJGLX0126。
文摘Accurate and efficient detection of building changes in remote sensing imagery is crucial for urban planning,disaster emergency response,and resource management.However,existing methods face challenges such as spectral similarity between buildings and backgrounds,sensor variations,and insufficient computational efficiency.To address these challenges,this paper proposes a novel Multi-scale Efficient Wavelet-based Change Detection Network(MewCDNet),which integrates the advantages of Convolutional Neural Networks and Transformers,balances computational costs,and achieves high-performance building change detection.The network employs EfficientNet-B4 as the backbone for hierarchical feature extraction,integrates multi-level feature maps through a multi-scale fusion strategy,and incorporates two key modules:Cross-temporal Difference Detection(CTDD)and Cross-scale Wavelet Refinement(CSWR).CTDD adopts a dual-branch architecture that combines pixel-wise differencing with semanticaware Euclidean distance weighting to enhance the distinction between true changes and background noise.CSWR integrates Haar-based Discrete Wavelet Transform with multi-head cross-attention mechanisms,enabling cross-scale feature fusion while significantly improving edge localization and suppressing spurious changes.Extensive experiments on four benchmark datasets demonstrate MewCDNet’s superiority over comparison methods:achieving F1 scores of 91.54%on LEVIR,93.70%on WHUCD,and 64.96%on S2Looking for building change detection.Furthermore,MewCDNet exhibits optimal performance on the multi-class⋅SYSU dataset(F1:82.71%),highlighting its exceptional generalization capability.
基金NR SEQUOIA(ANR-22-CE18-0006)Nan0Gold(ANR-22-CE29-0022)+3 种基金SAMURAI(ANR-24-CE19-2073-01)Wilive(ANR-24-CE09-2351-03)EUR CBH-EUR-GS(ANR-17-EURE 0003)for their financial supportthe French National Research Agency(Labex ARCANE,ANR-11-LABX-003 and CBH-EUR-GS,ANR-17-EURE-0003)that supported part of this study.
文摘Gold nanoclusters(AuNCs)are ultrasmall(<2 nm)aggregates of gold atoms that exhibit discrete electronic states,size-dependent photoluminescence,and exceptional biocompatibility,making them ideal candidates for theranostic applications.Their tunable surface chemistry enables targeted delivery,while strong near-infrared emission and environmental responsiveness allow for sensitive detection and deep-tissue imaging.Recent advances have revealed that controlled assembly of AuNCs into higher-order architectures-guided by biological scaffolds such as nucleic acids,peptides,and proteins-can markedly enhance their optical and electronic properties through aggregation-induced emission(AiE)and stabilization of surface ligands.This review summarizes recent progress in the design and biomedical applications of AuNC assemblies generated using biomolecules as structure-directing scaffolds.Covalent and noncovalent interactions with biomolecules enable the formation of well-defined one-,two-,and three-dimensional structures with tunable morphologies and sizes.These assemblies display distinctive photophysical behaviors that have been exploited for biosensing,bioimaging,and therapeutic applications in both cellular and in vivo models.Compared with individual AuNCs,assembled systems offer improved uptake,prolonged circulation,and efficient clearance,while protecting labile cargos such as nucleic acids and proteins.Moreover,their ordered and defined architectures can be engineered for controlled drug release and synergistic photo-or radiotherapeutic effects.Despite these advances,fundamental understanding of how structural organization governs photophysical responses remains limited.Elucidating parameters such as intercluster spacing and loading density will be essential for optimizing performance.Overall,biologically guided AuNC assemblies represent a powerful platform for multifunctional biosensing and therapy,bridging nanoscale design with biological function.
基金Tianmin Tianyuan Boutique Vegetable Industry Technology Service Station(Grant No.2024120011003081)Development of Environmental Monitoring and Traceability System for Wuqing Agricultural Production Areas(Grant No.2024120011001866)。
文摘Tomato is a major economic crop worldwide,and diseases on tomato leaves can significantly reduce both yield and quality.Traditional manual inspection is inefficient and highly subjective,making it difficult to meet the requirements of early disease identification in complex natural environments.To address this issue,this study proposes an improved YOLO11-based model,YOLO-SPDNet(Scale Sequence Fusion,Position-Channel Attention,and Dual Enhancement Network).The model integrates the SEAM(Self-Ensembling Attention Mechanism)semantic enhancement module,the MLCA(Mixed Local Channel Attention)lightweight attention mechanism,and the SPA(Scale-Position-Detail Awareness)module composed of SSFF(Scale Sequence Feature Fusion),TFE(Triple Feature Encoding),and CPAM(Channel and Position Attention Mechanism).These enhancements strengthen fine-grained lesion detection while maintaining model lightweightness.Experimental results show that YOLO-SPDNet achieves an accuracy of 91.8%,a recall of 86.5%,and an mAP@0.5 of 90.6%on the test set,with a computational complexity of 12.5 GFLOPs.Furthermore,the model reaches a real-time inference speed of 987 FPS,making it suitable for deployment on mobile agricultural terminals and online monitoring systems.Comparative analysis and ablation studies further validate the reliability and practical applicability of the proposed model in complex natural scenes.
文摘Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posing a significant challenge for accurate low-light object detection.Existing methods struggle to preserve discriminative features while maintaining semantic consistency between low-light and normal-light images.For this purpose,this study proposes a DL-YOLO model specially tailored for low-light detection.To mitigate target feature attenuation introduced by repeated downsampling,we design aMulti-Scale FeatureConvolution(MSF-Conv)module that captures rich,multi-level details via multi-scale feature learning,thereby reducing model complexity and computational cost.For feature fusion,we integrated the C3k2-DWRmodule by embedding the Dilation-wise Residual(DWR)mechanism into the 2-core optimized Cross Stage Partial(C3)framework,achieving efficient feature integration.In addition,we replace conventional localization losses with WIoU(Weighted Intersection over Union),which dynamically adjusts gradient gain according to sample quality,thereby improving localization robustness and precision.Experiments on the ExDark dataset demonstrate that DL-YOLO delivers strong low-light detection performance.The relevant code is published at https://github.com/cym0997/DL-YOLO.
基金supported by National Natural Science Foundation of China(Grant 22407024)the Star-up Research Fund of Southeast University(RF1028624094)(X.W.)+7 种基金the China Postdoctoral Science Foundation(Grant 2025M772911)Natural Science Foundation of Jiangsu Province(Grants BK20251303)(X.L.)Postdoctoral Fellowship Program of CPSF(Grant GZC20251914)(X.L.)Jiangsu Funding Program for Excellent Postdoctoral Talent(Grant 2025ZB052)(X.L.)National Natural Science Foundation of China(Grant 22234002)(G.L.)National Key Research and Development Program of China(Grant 2023YFF0724100)(G.L.)Natural Science Foundation of Jiangsu Province(Grant BK20232007)(G.L.)Jiangsu ShuangChuang Team(Grant JSSCTD202409)(G.L.and X.W.).
文摘Targeted protein degradation(TPD)is an innovative strategy for selectively eliminating pathogenic proteins,enabling precise degradation of once-undruggable targets in cancer therapy.However,current TPD molecules are often limited by poor tumor targeting and the need for high doses.To overcome these limitations,assembly/disassembly-based TPD systems have been proposed to effectively degrade proteins of interest and enhance therapeutic efficacy.Herein,we summarize the recent advances in such TPD systems and categorize the strategies employed,including nanosphere morphology of assembled TPD systems,nanofiber morphology of assembled TPD systems,carrier-mediated TPD release systems,and stimulus-induced free TPD molecule formation nanosystems.Finally,we outline future directions and identify the remaining challenges in assembly/disassembly-based TPD systems.
基金Supported by the National Natural Science Foundation of China(Nos.42141003,42176147)the National Key Research and Development Program of China(No.2022YFF0802204)the Xiamen Key Laboratory of Urban Sea Ecological Conservation and Restoration(USER)(Nos.USER2021-1,USER2021-5)。
文摘Ecological floating bed is an important biological remediation method for water pollution control.During the removal of excess nutrients and pollutants,changes in environmental factors affect the characteristics of microorganisms in aquatic ecosystems.To understand the influences of ecological floating beds on size-fractionated microorganisms,we investigated the community assembly and nitrogen metabolic characteristics of three size-fractionated microorganism groups in the ecological floating bed area,using 18S rDNA,16S rDNA metabarcoding,and metagenomic sequencing techniques.Firstly,we discovered substantial differences between size-fractionated groups in the diversity and compositions of both microeukaryotic and bacterial communities,as well as the influences of floating beds on specific groups.The floating beds appeared to provide more habitats for heterotrophs and symbiotes while potentially inhibiting the growth of certain phytoplankton(cyanobacteria).Secondly,we observed that microeukaryotic and bacterial communities were predominantly influenced by stochastic and deterministic processes,respectively,and they both exhibited distinct patterns across different size-fractionated groups.Notably,microeukaryotic community assembly demonstrated a greater sensitivity to ecological floating beds,as indicated by an increase in dispersal limitation processes.Finally,the nitrogen metabolism functional genes revealed that microbes associated with large-sized particles played a crucial role in dissimilatory nitrate reduction to ammonium(DNRA)and denitrification processes within the floating bed area,thereby facilitating the removal of excess nitrogen nutrients from the water.In contrast,freeliving microorganisms from small-sized groups were linked mainly to the genes involved in nitrogen assimilation and assimilatory nitrate reduction to ammonium(ANRA)processes.These findings help understand the impact of ecological floating beds on the diversity and functional characteristics of microorganism communities in different size-fractionated groups.
基金supported by the National Key Research and Development of China(No.2022YFB2503400).
文摘Semantic segmentation for mixed scenes of aerial remote sensing and road traffic is one of the key technologies for visual perception of flying cars.The State-of-the-Art(SOTA)semantic segmentation methods have made remarkable achievements in both fine-grained segmentation and real-time performance.However,when faced with the huge differences in scale and semantic categories brought about by the mixed scenes of aerial remote sensing and road traffic,they still face great challenges and there is little related research.Addressing the above issue,this paper proposes a semantic segmentation model specifically for mixed datasets of aerial remote sensing and road traffic scenes.First,a novel decoding-recoding multi-scale feature iterative refinement structure is proposed,which utilizes the re-integration and continuous enhancement of multi-scale information to effectively deal with the huge scale differences between cross-domain scenes,while using a fully convolutional structure to ensure the lightweight and real-time requirements.Second,a welldesigned cross-window attention mechanism combined with a global information integration decoding block forms an enhanced global context perception,which can effectively capture the long-range dependencies and multi-scale global context information of different scenes,thereby achieving fine-grained semantic segmentation.The proposed method is tested on a large-scale mixed dataset of aerial remote sensing and road traffic scenes.The results confirm that it can effectively deal with the problem of large-scale differences in cross-domain scenes.Its segmentation accuracy surpasses that of the SOTA methods,which meets the real-time requirements.
基金National Key Research&Development Program of China,No.2021YFC3201201Ningxia Key Research and Development Program(Special Talents),No.2023BSB03021+1 种基金Natural Science Foundation of Ningxia,No.2023AAC05014University First-Class Discipline Construction Project of Ningxia,No.NXYLXK2021A03。
文摘The Qingtongxia Irrigation District in Ningxia is an important hydrological and ecological region.To assess its ecological environment quality from 2001 to 2021 across multiple scales and identify driving factors,a modified remote sensing ecological index(MRSEI)was developed by incorporating evapotranspiration.Spatial and temporal patterns were analyzed using the coefficient of variation,spatial autocorrelation,and semi-variogram methods,while influencing factors were explored via the optimal parameter geographical detector model.The MRSEI’s first principal component loadings and rankings aligned with those of RSEI(average contribution:81.31%),effectively reflecting spatiotemporal variations.At sub-irrigation district and landscape scales,ecological quality was slightly lower than at the district level but remained stable.Moderate and good ecological grades accounted for 36.28%and 33.38%of the area,respectively,at the district scale,and the moderate grade reached 70.48%on smaller scales.Spatial heterogeneity intensified with decreasing scale,and human activity lost explanatory power below a 5 km range.Human factors mainly drove ecological differentiation at the district scale,while natural factors dominated at finer scales.The MRSEI offers a novel tool for ecological assessment in arid/semi-arid areas and supports scale-adapted ecological protection strategies.
文摘Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.
基金financial support from the National Key Research and Development Program of China(2022YFB3804905)National Natural Science Foundation of China(22375047,22378068,and 22378071)+1 种基金Natural Science Foundation of Fujian Province(2022J01568)111 Project(No.D17005).
文摘Advanced healthcare monitors for air pollution applications pose a significant challenge in achieving a balance between high-performance filtration and multifunctional smart integration.Electrospinning triboelectric nanogenerators(TENG)provide a significant potential for use under such difficult circumstances.We have successfully constructed a high-performance TENG utilizing a novel multi-scale nanofiber architecture.Nylon 66(PA66)and chitosan quaternary ammonium salt(HACC)composites were prepared by electrospinning,and PA66/H multiscale nanofiber membranes composed of nanofibers(≈73 nm)and submicron-fibers(≈123 nm)were formed.PA66/H multi-scale nanofiber membrane as the positive electrode and negative electrode-spun PVDF-HFP nanofiber membrane composed of respiration-driven PVDF-HFP@PA66/H TENG.The resulting PVDF-HFP@PA66/H TENG based air filter utilizes electrostatic adsorption and physical interception mechanisms,achieving PM_(0.3)filtration efficiency over 99%with a pressure drop of only 48 Pa.Besides,PVDF-HFP@PA66/H TENG exhibits excellent stability in high-humidity environments,with filtration efficiency reduced by less than 1%.At the same time,the TENG achieves periodic contact separation through breathing drive to achieve self-power,which can ensure the long-term stability of the filtration efficiency.In addition to the air filtration function,TENG can also monitor health in real time by capturing human breathing signals without external power supply.This integrated system combines high-efficiency air filtration,self-powered operation,and health monitoring,presenting an innovative solution for air purification,smart protective equipment,and portable health monitoring.These findings highlight the potential of this technology for diverse applications,offering a promising direction for advancing multifunctional air filtration systems.
文摘Defect detection in printed circuit boards(PCB)remains challenging due to the difficulty of identifying small-scale defects,the inefficiency of conventional approaches,and the interference from complex backgrounds.To address these issues,this paper proposes SIM-Net,an enhanced detection framework derived from YOLOv11.The model integrates SPDConv to preserve fine-grained features for small object detection,introduces a novel convolutional partial attention module(C2PAM)to suppress redundant background information and highlight salient regions,and employs a multi-scale fusion network(MFN)with a multi-grain contextual module(MGCT)to strengthen contextual representation and accelerate inference.Experimental evaluations demonstrate that SIM-Net achieves 92.4%mAP,92%accuracy,and 89.4%recall with an inference speed of 75.1 FPS,outperforming existing state-of-the-art methods.These results confirm the robustness and real-time applicability of SIM-Net for PCB defect inspection.
基金financially supported by the National Natural Science Foundation of China(Nos.42272137,42202160)AAPG Foundation Grants-in-Aid and the Strategic Cooperation Technology Projecti of CNPC and CUPB(No.ZLZX2020-01-05)。
文摘The pore structure of shale oil reservoir significantly affects the occurrence and mobility of hydrocarbons.The potential of a new type of alkaline lake shale oil has been demonstrated,but there are few reports on the pore system of alkaline lake shale,which restricts the efficient exploration and development of shale oil.This study investigates the Fengcheng Formation shale in the Mahu sag of the Junggar Basin,employing methods such as low-temperature nitrogecn adsorption(LTNA),mercury intrusion capillary pressure(MICP),and nuclear magnetic resonance(NMR)to quantitatively characterize the multi-scale pore structure and fractal characteristics of shale,while evaluating the applicability of these methods.Based on a comprehensive analysis of material composition,different pore types,and fractal dimensions,the controlling factors for the development of different pore types and their seepage capacity are discussed.The results indicate that inorganic mineral pores are the main development in alkaline lake shale,with the pore morphology being characterized by slit-like and ink-bottle shapes.The multi-scale pore size distribution(PSD)shows that Ⅱ-micropores(10-100 nm)and mesopores(100–1000 nm)are the main contributors to the pore system.The development of Ⅱ-micropores is associated with feldspar and calcareous minerals,the development of Ⅰ-micropores(<10 nm)and mesopores is related to quartz content,while large pores are mainly found in interlayer fissures of clay minerals.The development of Ⅰ-micropores increases the roughness of pore surface and enhances the adsorption capacity of the pores,while the development of Ⅱ-micropores associated with calcareous minerals hinders pore seepage capacity.Mesopores and macropores(>1000 nm)exhibit good flowability.The high content of siliceous minerals plays a positive role in the pore system of alkaline lake shale.The shale with higher fractal dimension Dmin exhibits greater adsorption capacity,which hinders the accumulation of free-state shale oil.Different types of pore space play different roles in the occurrence of shale oil,with free-state shale oil primarily occurring in micro-fractures and inorganic mineral pores,and the pore size is exceeding 10 nm.
基金supported by Key Laboratory of Cyberspace Security,Ministry of Education,China。
文摘Transformer-based models have significantly advanced binary code similarity detection(BCSD)by leveraging their semantic encoding capabilities for efficient function matching across diverse compilation settings.Although adversarial examples can strategically undermine the accuracy of BCSD models and protect critical code,existing techniques predominantly depend on inserting artificial instructions,which incur high computational costs and offer limited diversity of perturbations.To address these limitations,we propose AIMA,a novel gradient-guided assembly instruction relocation method.Our method decouples the detection model into tokenization,embedding,and encoding layers to enable efficient gradient computation.Since token IDs of instructions are discrete and nondifferentiable,we compute gradients in the continuous embedding space to evaluate the influence of each token.The most critical tokens are identified by calculating the L2 norm of their embedding gradients.We then establish a mapping between instructions and their corresponding tokens to aggregate token-level importance into instructionlevel significance.To maximize adversarial impact,a sliding window algorithm selects the most influential contiguous segments for relocation,ensuring optimal perturbation with minimal length.This approach efficiently locates critical code regions without expensive search operations.The selected segments are relocated outside their original function boundaries via a jump mechanism,which preserves runtime control flow and functionality while introducing“deletion”effects in the static instruction sequence.Extensive experiments show that AIMA reduces similarity scores by up to 35.8%in state-of-the-art BCSD models.When incorporated into training data,it also enhances model robustness,achieving a 5.9%improvement in AUROC.
基金financial y supported by the National Key Research and Development Program of China (2023YFD1900902)the Joint Funds of the Zhejiang Provincial Natural Science Foundation of China (LLSSZ24C030001)+1 种基金the Earmarked Fund for China Agriculture Research System (CARS-08-G-09)sponsored by the K.C.Wong Magna Fund of Ningbo University,China。
文摘Fungi play crucial roles in nutrient acquisition,plant growth promotion,and the enhancement of resistance to both abiotic and biotic stresses.However,studies on the fungal communities associated with peas (Pisum sativum L.) remain limited.In this study,we systematically investigated the ecological effects of host niches (soil,root,stem,leaf,and pod) and genotypes on the diversity and composition of fungal communities in peas using a multi-level approach that encompassed pattern recognition (β-diversity decomposition),mechanism validation (neutral community model testing),and dynamic tracking methods (migration pathway source-tracking).The results revealed that the dominant fungal phyla across niches and genotypes were Ascomycota,Basidiomycota,and Mortierellomycota,and the community structures of the soil–plant continuum were primarily determined by the pea niches rather than genotypes.β-diversity decomposition was largely attributed to species replacement rather than richness differences,indicating strong niche specificity and microbial replacement across microhabitats.Neutral model analysis revealed that stochastic processes influenced genotypeassociated communities,while deterministic processes played a dominant role in niche-based community assembly.Source-tracking analysis identified niche-to-niche fungal migration,with Erysiphe,Fusarium,Cephaliophora,Ascobolus,Alternaria,and Aspergillus as the key genera.Migration rates from exogenous to endogenous niches were low (1.3–61.5%),whereas those within exogenous (64.4–83.7%) or endogenous (73.9–96.4%) compartments were much higher,suggesting that the pea epidermis acts as a selective barrier that filters and enriches microbial communities prior to internal colonization.This study provides comprehensive insights into the mechanisms of host filtering,enrichment and microbial sourcing,which increases our understanding of the assembly rules of the pea-associated fungal microbiome.
