Surface/underwater target classification is a key topic in marine information research.However,the complex underwater environment,coupled with the diversity of target types and their variable characteristics,presents ...Surface/underwater target classification is a key topic in marine information research.However,the complex underwater environment,coupled with the diversity of target types and their variable characteristics,presents significant challenges for classifier design.For shallow-water waveguides with a negative thermocline,a residual neural network(ResNet)model based on the sound field elevation structure is constructed.This model demonstrates robust classification performance even when facing low signal-to-noise ratios and environmental mismatches.Meanwhile,to address the reduced generalization ability caused by limited labeled acoustic data,an improved ResNet model based on unsupervised domain adaptation(“proposed UDA-ResNet”)is further constructed.This model incorporates data on simulated elevation structures of the sound field to augment the training process.Adversarial training is employed to extract domain-invariant features from simulated and trial data.These strategies help reduce the negative impact caused by domain differences.Experimental results demonstrate that the proposed method shows strong surface/underwater target classification ability under limited sample sizes,thus confirming its feasibility and effectiveness.展开更多
This paper focuses on the application of the adaptive mesh method in the study of underwater shock waves near the water surface.By integrating theoretical analysis with a five-equation model under axisymmetric coordin...This paper focuses on the application of the adaptive mesh method in the study of underwater shock waves near the water surface.By integrating theoretical analysis with a five-equation model under axisymmetric coordinates,we developed an optimized computational framework for multi-material fluid simulations.The moving mesh method(r-method)is used to accurately capture complex underwater shock wave systems.Multiple numerical experiments are conducted,including deep-water explosions,near-surface explosions for both spherical charge and cylindrical charge,and regular–irregular reflection interface calculations.The results show that compared to the fixed mesh method,the adaptive mesh method provides results closer to the theoretical values and achieves local high-resolution computation of multi-material fluids.By adjusting the adaptive function,different mesh refinement effects can be obtained.This method also has certain advantages in calculating the regular–irregular reflection interface in underwater explosions.This study establishes a validated computational framework for underwater explosion research,enhancing the predictive accuracy for underwater shock wave propagation in engineering assessments and providing new insights into the fundamental physics of multi-material explosion phenomena.展开更多
Biomass-derived carbon materials are favored for their abundance and sustainability,and ease of preparation and modification.By surface activation and modification they can have a good electrical conductivity,excellen...Biomass-derived carbon materials are favored for their abundance and sustainability,and ease of preparation and modification.By surface activation and modification they can have a good electrical conductivity,excellent catalytic activity,a remarkable adsorption capacity,and different interfacial physicochemical functionalities.Surface-modified biochars have found wide applications in energy storage,environmental remediation,and catalysis.However,achieving precise and controllable modification of their active sites remains a challenge.Recent advances and future prospects for controlling their surface morphology,defect engineering,and surface coating strategies,with particular attention to their means of fabrication,are reviewed.展开更多
The continuous decrease in global fishery resources has increased the importance of precise and efficient underwater fish monitoring technology.First,this study proposes an improved underwater target detection framewo...The continuous decrease in global fishery resources has increased the importance of precise and efficient underwater fish monitoring technology.First,this study proposes an improved underwater target detection framework based on YOLOv8,with the aim of enhancing detection accuracy and the ability to recognize multi-scale targets in blurry and complex underwater environments.A streamlined Vision Transformer(ViT)model is used as the feature extraction backbone,which retains global self-attention feature extraction and accelerates training efficiency.In addition,a detection head named Dynamic Head(DyHead)is introduced,which enhances the efficiency of processing various target sizes through multi-scale feature fusion and adaptive attention modules.Furthermore,a dynamic loss function adjustment method called SlideLoss is employed.This method utilizes sliding window technology to adaptively adjust parameters,which optimizes the detection of challenging targets.The experimental results on the RUOD dataset show that the proposed improved model not only significantly enhances the accuracy of target detection but also increases the efficiency of target detection.展开更多
Hydrogen peroxide(H_(2)O_(2))oxidation and reduction reactions(HPOR/HPRR)are pivotal in various innovative electrochemical energy conversion devices.A comprehensive understanding of these mechanisms is critical for ca...Hydrogen peroxide(H_(2)O_(2))oxidation and reduction reactions(HPOR/HPRR)are pivotal in various innovative electrochemical energy conversion devices.A comprehensive understanding of these mechanisms is critical for catalyst design and performance improvement in these applications.In this work,we systematically investigate the HPOR/HPRR mechanisms on low-index Pt surfaces,specifically Pt(111),Pt(100)and Pt(110),through density functional theory(DFT)calculations combined with the computational hydrogen electrode(CHE)model.For HPOR,all the low-index Pt surfaces exhibit a unified potential-determining step(PDS)involving the electrochemical oxidation of hydroperoxyl intermediates(HOO*).The binding free energy of HOO*(Δ_(GHOO*))emerges as an activity descriptor,with Pt(110)exhibiting the highest HPOR activity.The HPRR mechanism follows a chem-electrochemical(C-EC)pathway.The rate-determining step(RDS)of HPRR is either the cleavage of the HO-OH bond(chemical)or the reduction of HO(electrochemical),depending on their respective activation energies.These activation energies are functions of the HO*binding free energy,Δ_(GHO*),establishingΔ_(GHO*)as the descriptor for HPRR activity prediction.Pt(111)and Pt(100)are identified as the most active HPRR catalysts among the studied metal surfaces,although they still experience a significant overpotential.The scaling relationship betweenΔ_(GHOO*)andΔ_(GHO*)reveals a thermodynamic coupling of HPOR and HPRR,explaining their occurrence on Pt surfaces.These findings provide important insights and activity descriptors for both HPOR and HPRR,providing valuable guidance for the design of electrocatalysts in H_(2)O_(2)-related energy applications and fuel cells.展开更多
Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challeng...Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.展开更多
Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet ...Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet architecture integrates Laplacian pyramid depth residuals and Sobel operators to improve the boundary details in depth images,which may suffer from the feature loss caused by upsampling and the blurriness of underwater images.Multiscale local planar guidance layers then fully exploit the intermediate depth features,and a comprehensive loss function ensures robustness and accuracy.Experimental results on benchmarks demonstrate the effectiveness of Lpg-Lap Unet and its superior performance over state-of-the-art models.An underwater target tracking system is then designed to further validate its real-time capabilities in the AirSim simulation platform.展开更多
This study investigated surface roughness,the wettability behavior,and surface energy of Co-based alloy specimens textured using the biomimetic Laser Surface Texturing(LST)method.The surface texture was inspired by th...This study investigated surface roughness,the wettability behavior,and surface energy of Co-based alloy specimens textured using the biomimetic Laser Surface Texturing(LST)method.The surface texture was inspired by the patterns found on marine shells.The impacts of the parameters on wettability,Surface Free Energy(SFE),surface topography,and texture roughness generated by the laser beam tracking a spiral path were investigated.Reducing spiral pitch produces more complicated and chaotic surface patterns.Most surfaces are hydrophobic,and surface roughness and topography influence the Contact Angle(CA).Topography and roughness were affected by frequency and scanning speed;a decrease in scanning speed and frequency generated more chaotic and irregular surface textures.With general factorial analysis and Analysis of Variance(ANOVA),our statistical study reveals that accounting for 88%of the influence,the scanning speed is the primary factor influencing surface roughness.On the other hand,the spiral pitch is essential for defining the struc-tural features of the surface,even if it less influences roughness.The SFE of laser-textured CoCr28Mo alloy specimens was optimizable within the range of 14-32 mN/m.The relevant findings offer valuable insights into optimizing LST for the specific surface properties of the Co-based alloy.展开更多
This study experimentally investigates the influence of surfacewettability on the frosting characteristics of three types of corrugated structures(Types A,B,and C)under controlled low-temperature conditions.The experi...This study experimentally investigates the influence of surfacewettability on the frosting characteristics of three types of corrugated structures(Types A,B,and C)under controlled low-temperature conditions.The experiments were conducted in a constant-temperature bath at a cold surface temperature of–5℃,relative humidity of 90%,and ambient air temperature of 10℃.The results reveal that the variation trends of frost morphology,frost mass,and frost layer thickness are generally consistent across surfaces with different wettability.Among the tested surfaces,frost crystal formation and complete surface coverage occurred latest on the superhydrophobic surface(CA=153.9–165.8℃),next on the bare aluminumsurface(75.3–83.2℃),and earliest on the hydrophilic surface(5.3–7.5℃).At the same frosting duration,the superhydrophobic surface exhibited a sparse and fluffy frost layer,the bare aluminum surface formed a rough and dense frost,while the hydrophilic surface developed a fine and compact frost layer.The amount of frost formation decreased in the order of hydrophilic>bare aluminum>superhydrophobic,indicating that the superhydrophobic surface provides the most significant anti-frosting effect during the initial stages of frost formation.For instance,on the Type A corrugated structure,after 15 min of frosting,the frost mass on the superhydrophobic surface was 38.78%and 68.45%lower than those on the bare aluminum and hydrophilic surfaces,respectively.After 30 min,these differences were 4.99%and 25.26%,respectively.Overall,the superhydrophobic surface exhibited the smallest frost mass and frost layer thickness,demonstrating superior anti-frosting performance compared with the other surface types.展开更多
In this investigation,a hybrid approach integrating the IDDES turbulence model and FW-H is employed to forecast the hydroacoustic of the rim driven thruster(RDT)under non-cavitation and uniform flow conditions at vary...In this investigation,a hybrid approach integrating the IDDES turbulence model and FW-H is employed to forecast the hydroacoustic of the rim driven thruster(RDT)under non-cavitation and uniform flow conditions at varying loading conditions(J=0.3 and J=0.6).It is revealed that the quadrupole term contribution in the P-FWH method significantly affects the monopole term in the low-frequency region,while it mainly affects the dipole term in the high-frequency region.Specifically,the overall sound pressure levels(SPL)of the RDT using the P-FWH method are 2.27 dB,10.03 dB,and 16.73 dB at the receiving points from R1 to R3 under the heavy-loaded condition,while they increase by 0.67 dB at R1,and decrease by 14.93 dB at R2,and 22.20 dB at R3,for the light-loaded condition.The study also utilizes the pressure-time derivatives to visualize the numerical noise and to pinpoint the dynamics of the vortex cores,and the optimization of the grid design can significantly reduce the numerical noise.The computational accuracy of the P-FWH method can meet the noise requirements for the preliminary design of rim driven thrusters.展开更多
Vortex-induced vibration(VIV)of an underwater manipulator in pulsating flow presents a notable engineering problem in precise control due to the velocity variation in the flow.This study investigates the VIV response ...Vortex-induced vibration(VIV)of an underwater manipulator in pulsating flow presents a notable engineering problem in precise control due to the velocity variation in the flow.This study investigates the VIV response of an underwater manipulator subjected to pulsating flow,focusing on how different postures affect the behavior of the system.The effects of pulsating parameters and manipulator arrangement on the hydrodynamic coefficient,vibration response,motion trajectory,and vortex shedding behaviors were analyzed.Results indicated that the cross flow vibration displacement in pulsating flow increased by 32.14%compared to uniform flow,inducing a shift in the motion trajectory from a crescent shape to a sideward vase shape.In the absence of interference between the upper and lower arms,the lift coefficient of the manipulator substantially increased with rising pulsating frequency,reaching a maximum increment of 67.0%.This increase in the lift coefficient led to a 67.05%rise in the vibration frequency of the manipulator in the in-line direction.As the pulsating amplitude increased,the drag coefficient of the underwater manipulator rose by 36.79%,but the vibration frequency in the cross-flow direction decreased by 56.26%.Additionally,when the upper and lower arms remained in a state of mutual interference,the cross-flow vibration amplitudes of the upper and lower arms were approximately 1.84 and 4.82 times higher in a circular-elliptical arrangement compared to an elliptical-circular arrangement,respectively.Consequently,the flow field shifted from a P+S pattern to a disordered pattern,disrupting the regularity of the motion trajectory.展开更多
Lithium-sulfur batteries(LSBs)represent a next-generation energy storage technology,but widespread applications are restricted by the shuttle of lithium polysulfides(LiPSs).The rational design of separators has been d...Lithium-sulfur batteries(LSBs)represent a next-generation energy storage technology,but widespread applications are restricted by the shuttle of lithium polysulfides(LiPSs).The rational design of separators has been demonstrated to be one of the most efficient and cost-effective strategies to curb the shuttle effect,and tremendous research progress has been achieved.The efficiency of a separator depends on its interaction with LiPSs,which is governed by the surface energy and binding strength.Despite several review works that have been reported to advance the separators,most of them primarily focus on active material innovation and construction.The most crucial issues of surface binding energy have not been systematically reviewed,limiting the precise design of efficient separators.In this review,fundamentals related to surface energy and binding interactions with LiPSs are comprehensively analyzed and discussed.With surface binding and energy main lines,the advancements in separator engineering strategies are elaborately summarized and discussed.Moreover,techniques for evaluating affinity to LiPSs are thoroughly analyzed to avoid any ambiguities in measurement.Based on the research context,valuable research directions are suggested to construct efficient separators.This work provides guidelines to regulate the surface binding and energy of separators for high-performance LSBs.展开更多
The specific surface area(S S)and pore size(D)exhibit an inherent trade-off in the microscale design of bone implants:larger pores typically correlate with reduced surface area and vice versa.This relationship has att...The specific surface area(S S)and pore size(D)exhibit an inherent trade-off in the microscale design of bone implants:larger pores typically correlate with reduced surface area and vice versa.This relationship has attracted notable attention because of its critical role in the regulation of cell adhesion and osteogenesis.However,it remains largely unclear how S S and D affect the generated bone tissue and dynamically change during long-term osteogenesis.Herein,by applying rigorous geometric mapping to minimal surfaces,we constructed precisely partitioned and layer-by-layer thickened tissue models to simulate osteogenesis across different temporal scales and thereby track the dynamic evolution of geometric characteristics,permeability,and mechanobiological tissue differentiation.The high-S S samples were found to facilitate the rapid formation of new bone tissue in the early stages.However,their smaller pores tended to cause occlusions,hindering further tissue development.In contrast,low-S S samples showed slower bone regeneration,but their larger pores provided adequate physical space for tissue regeneration and mass transport,ultimately promoting bone formation in the long term.Mechanobiological regulation suggests that fibrous tissue formation inhibits additional bone formation,establishing a dynamic equilibrium between osteogenesis and pore space to sustain nutrient/waste exchange throughout the regenerative process.Overall,smaller pores are preferable in implants for minimally loaded osteoplasty procedures focused on early-stage bone consolidation,whereas larger pores are preferable in dynamically loaded implants requiring prolonged mechanical stability.展开更多
The original online version of this article was revised:Several errors occurred in the published version of the article.These have now been corrected as follows:Page 2,section"2.2 Laser Texturing Procedure of Sur...The original online version of this article was revised:Several errors occurred in the published version of the article.These have now been corrected as follows:Page 2,section"2.2 Laser Texturing Procedure of Surfaces",line 2:The device name was corrected from"YDFLP-E-50-M8"to"YDFLP-50-M8."Page 3,Section 2.4:The phrase"95%confidence interval"has been corrected to"95%confidence level."Page 3,Figure 1 caption:The phrase"fandg"has been corrected to"f and g."The order"C4 and C12"has been reversed to"C12 and C4,"in accordance with the display order in the figure.Page 4,Figure reference:The phrase"Figs.4c and d"has been corrected to"Figs.5b and c."Page 5,paragraph starting with"The ANOVA results are presented...":The phrase"95%confidence interval"has been corrected to"95%confidence level."展开更多
Photodetectors can convert light energy into electrical signals,so are widely used in photovoltaics,photon counting,monitoring,and imaging.Photodetectors are easy to prepare high-resolution photochips because of their...Photodetectors can convert light energy into electrical signals,so are widely used in photovoltaics,photon counting,monitoring,and imaging.Photodetectors are easy to prepare high-resolution photochips because of their small size unit integration.However,these photodetector units often exhibit poor photoelectric performance due to material defects and inadequate structures,which greatly limit the functions of devices.Designing modification strategies and micro-/nanostructures can compensate for defects,adjust the bandgap,and develop novel quantum structures,which consequently optimize photovoltaic units and revolutionize optoelectronic devices.Here,this paper aims to comprehensively elaborate on the surface/interface engineering scheme of micro-/nano-photodetectors.It starts from the fundamentals of photodetectors,such as principles,types,and parameters,and describes the influence of material selection,manufacturing techniques,and post-processing.Then,we analyse in detail the great influence of surface/interface engineering on the performance of photovoltaic devices,including surface/interface modification and micro-/nanostructural design.Finally,the applications and prospects of optoelectronic devices in various fields such as miniaturization of electronic devices,robotics,and human–computer interaction are shown.展开更多
Cascading failures pose a serious threat to the survivability of underwater unmanned swarm networks(UUSNs),significantly limiting their service ability in collaborative missions such as military reconnaissance and env...Cascading failures pose a serious threat to the survivability of underwater unmanned swarm networks(UUSNs),significantly limiting their service ability in collaborative missions such as military reconnaissance and environmental monitoring.Existing failure models primarily focus on power grids and traffic systems,and don't address the unique challenges of weak-communication UUSNs.In UUSNs,cascading failure present a complex and dynamic process driven by the coupling of unstable acoustic channels,passive node drift,adversarial attacks,and network heterogeneity.To address these challenges,a directed weighted graph model of UUSNs is first developed,in which node positions are updated according to ocean-current-driven drift and link weights reflect the probability of successful acoustic transmission.Building on this UUSNs graph model,a cascading failure model is proposed that integrates a normal-failure-recovery state-cycle mechanism,multiple attack strategies,and routingbased load redistribution.Finally,under a five-level connectivity UUSNs scheme,simulations are conducted to analyze how dynamic topology,network load,node recovery delay,and attack modes jointly affect network survivability.The main findings are:(1)moderate node drift can improve survivability by activating weak links;(2)based-energy routing(BER)outperform based-depth routing(BDR)in harsh conditions;(3)node self-recovery time is critical to network survivability;(4)traditional degree-based critical node metrics are inadequate for weak-communication UUSNs.These results provide a theoretical foundation for designing robust survivability mechanisms in weak-communication UUSNs.展开更多
To solve the problem of abnormal abrasion of Cu-Based Friction Materials(CBFMs),Bionic Non-Smooth Surface(BNS)on friction surface of CBFMs was constructed based on bionic principles,and the optimal bionic prototype wa...To solve the problem of abnormal abrasion of Cu-Based Friction Materials(CBFMs),Bionic Non-Smooth Surface(BNS)on friction surface of CBFMs was constructed based on bionic principles,and the optimal bionic prototype was selected by Finite Element Method(FEM).In addition,the bionic parameters were optimized by Response Surface Method(RSM).Samples holding BNS were prepared by Laser Processing,tribological properties were tested by a Friction and Wear Tester and worn surface morphology was characterized by a Scanning Electron Microscope(SEM).The results showed that BNS on friction surface could regulate the stress distribution and alleviate the peak stress.Among all samples,the coupled texture of pit-hexagonal got the minimum peak stress.During braking,bionic texture could also collect wear debris or change the motion forms from sliding to rotation,which can reduce abnormal abrasion.The wear rate was reduced by 19.31%.The results in this paper can provide a new idea for enhancing the tribological properties of CBFMs,and can also lay the foundation for further research of bionic tribology.展开更多
In this study,polyacrylic acid(PAA)films were employed as a model system,and a series of PAA films with tunable water wettability was systematically prepared by varying molecular weight and curing temperature.Using at...In this study,polyacrylic acid(PAA)films were employed as a model system,and a series of PAA films with tunable water wettability was systematically prepared by varying molecular weight and curing temperature.Using attenuated total reflectance Fourier-transform infrared spectroscopy(ATR-FTIR),the molecular configurations of surface carboxyl groups(COOH),free carboxyl(COOH_(f))and hydrogen-bonded carboxyl(COOH_(HB),were directly correlated with the polar component of surface energy(γ^(s,p)).By decomposing theγ^(s,p)values of the PAA thin films as a sum of the contributions of COOH_(f)and COOH_(H B),the intrinsic polar component of surface energy of COOH_(H B)(γ_(H B)^(s,p*))was quantified for the first time as 8.34 mN/m,significantly lower than that of COOH_(f)(γ_(f)^(s,p*)=34 mN/m).This result highlights that hydrogen bonding markedly reduces theγ^(s,p),providing a rational explanation for the relatively large water contact angle observed on PAA thin films.Furthermore,it establishes a thermodynamic basis for estimating the fraction of surface COOH_(H B)groups(f H B)from wettability measurements.Further extension of the model to carboxyl-terminated self-assembled monolayers(COOH-SAMs)revealed that surface COOH density(ΣCOOH)critically regulates wetting behavior:whenΣCOOH ranges from 4.30 to 5.25 nm^(-2),COOH groups predominantly exist in a free state and facilitate effective hydration layers,thereby promoting superhydrophilicity.Overall,this study not only establishes a unified thermodynamic framework linking surface COOH configurations to macroscopic wettability,but also validates its universality by extending it to COOH-SAMs systems,thereby providing a unified theoretical framework for the controllable design of hydrophilicity in various COOH-functionalized surfaces.展开更多
A methodology for the reduction of radar cross section(RCS)of cambered platforms within the target airspace is presented,which utilizes a dual-polarized ultra-wide-angle artificial electromagnetic absorbing surface.By...A methodology for the reduction of radar cross section(RCS)of cambered platforms within the target airspace is presented,which utilizes a dual-polarized ultra-wide-angle artificial electromagnetic absorbing surface.By applying the theory of generalized Brewster complex wave impedance matching,five distinct unit cell designs are developed to attain more than95%absorption rate for dual-polarized incident waves within five angular ranges:0°-30°,30°-50°,50°-60°,60°-70°,and 70°-80°.To optimally reduce the RCS of a cambered platform,the five types of units can be evenly distributed on the surface based on the local incident angles of plane waves originating from the target airspace.As an illustrative example,the leading edge of an airfoil is taken into account,and experimental measurements validate the efficiency of the proposed structure.Specifically,the absorbing surface achieves more than 10 dB of RCS reduction in the frequency ranges from 5-10 GHz(about66.7%relative bandwidth)for dual polarizations.展开更多
In industrial manufacturing,efficient surface defect detection is crucial for ensuring product quality and production safety.Traditional inspectionmethods are often slow,subjective,and prone to errors,while classicalm...In industrial manufacturing,efficient surface defect detection is crucial for ensuring product quality and production safety.Traditional inspectionmethods are often slow,subjective,and prone to errors,while classicalmachine vision techniques strugglewith complex backgrounds and small defects.To address these challenges,this study proposes an improved YOLOv11 model for detecting defects on hot-rolled steel strips using the NEU-DET dataset.Three key improvements are introduced in the proposed model.First,a lightweight Guided Attention Feature Module(GAFM)is incorporated to enhance multi-scale feature fusion,allowing the model to better capture and integrate semantic and spatial information across different layers,which improves its ability to detect defects of varying sizes.Second,an Aggregated Attention(AA)mechanism is employed to strengthen the representation of critical defect features while effectively suppressing irrelevant background information,particularly enhancing the detection of small,low-contrast,or complex defects.Third,Ghost Dynamic Convolution(GDC)is applied to reduce computational cost by generating low-cost ghost features and dynamically reweighting convolutional kernels,enabling faster inference without sacrificing feature quality or detection accuracy.Extensive experiments demonstrate that the proposed model achieves a mean Average Precision(mAP)of 87.2%,compared to 81.5%for the baseline,while lowering computational cost from6.3Giga Floating-point Operations Per Second(GFLOPs)to 5.1 GFLOPs.These results indicate that the improved YOLOv11 is both accurate and computationally efficient,making it suitable for real-time industrial surface defect detection and contributing to the development of practical,high-performance inspection systems.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.62471024 and 62301183)the Open Research Fund of Hanjiang Laboratory(KF2024001).
文摘Surface/underwater target classification is a key topic in marine information research.However,the complex underwater environment,coupled with the diversity of target types and their variable characteristics,presents significant challenges for classifier design.For shallow-water waveguides with a negative thermocline,a residual neural network(ResNet)model based on the sound field elevation structure is constructed.This model demonstrates robust classification performance even when facing low signal-to-noise ratios and environmental mismatches.Meanwhile,to address the reduced generalization ability caused by limited labeled acoustic data,an improved ResNet model based on unsupervised domain adaptation(“proposed UDA-ResNet”)is further constructed.This model incorporates data on simulated elevation structures of the sound field to augment the training process.Adversarial training is employed to extract domain-invariant features from simulated and trial data.These strategies help reduce the negative impact caused by domain differences.Experimental results demonstrate that the proposed method shows strong surface/underwater target classification ability under limited sample sizes,thus confirming its feasibility and effectiveness.
基金supported by the Overall Planning and Development Project of China Academy of Engineering Physics(CAEP)(Grant No.TCGH0909)the Science and Technology Development Fund of the Chinese Academy of Meteorological Sciences(CAMS)(Grant No.2023KJ048)the Basic Research Fund of the Chinese Academy of Meteorological Sciences(Grant No.2023Z001)。
文摘This paper focuses on the application of the adaptive mesh method in the study of underwater shock waves near the water surface.By integrating theoretical analysis with a five-equation model under axisymmetric coordinates,we developed an optimized computational framework for multi-material fluid simulations.The moving mesh method(r-method)is used to accurately capture complex underwater shock wave systems.Multiple numerical experiments are conducted,including deep-water explosions,near-surface explosions for both spherical charge and cylindrical charge,and regular–irregular reflection interface calculations.The results show that compared to the fixed mesh method,the adaptive mesh method provides results closer to the theoretical values and achieves local high-resolution computation of multi-material fluids.By adjusting the adaptive function,different mesh refinement effects can be obtained.This method also has certain advantages in calculating the regular–irregular reflection interface in underwater explosions.This study establishes a validated computational framework for underwater explosion research,enhancing the predictive accuracy for underwater shock wave propagation in engineering assessments and providing new insights into the fundamental physics of multi-material explosion phenomena.
文摘Biomass-derived carbon materials are favored for their abundance and sustainability,and ease of preparation and modification.By surface activation and modification they can have a good electrical conductivity,excellent catalytic activity,a remarkable adsorption capacity,and different interfacial physicochemical functionalities.Surface-modified biochars have found wide applications in energy storage,environmental remediation,and catalysis.However,achieving precise and controllable modification of their active sites remains a challenge.Recent advances and future prospects for controlling their surface morphology,defect engineering,and surface coating strategies,with particular attention to their means of fabrication,are reviewed.
基金supported by the National Natural Science Foundation of China(No.52106080)the Jilin City Science and Technology Innovation Development Plan Project(No.20240302014)+2 种基金the Jilin Provincial Department of Education Science and Technology Research Project(No.JJKH20230135K)the Jilin Province Science and Technology Development Plan Project(No.YDZJ202401640ZYTS)the Northeast Electric Power University Teaching Reform Research Project(No.J2427)。
文摘The continuous decrease in global fishery resources has increased the importance of precise and efficient underwater fish monitoring technology.First,this study proposes an improved underwater target detection framework based on YOLOv8,with the aim of enhancing detection accuracy and the ability to recognize multi-scale targets in blurry and complex underwater environments.A streamlined Vision Transformer(ViT)model is used as the feature extraction backbone,which retains global self-attention feature extraction and accelerates training efficiency.In addition,a detection head named Dynamic Head(DyHead)is introduced,which enhances the efficiency of processing various target sizes through multi-scale feature fusion and adaptive attention modules.Furthermore,a dynamic loss function adjustment method called SlideLoss is employed.This method utilizes sliding window technology to adaptively adjust parameters,which optimizes the detection of challenging targets.The experimental results on the RUOD dataset show that the proposed improved model not only significantly enhances the accuracy of target detection but also increases the efficiency of target detection.
基金Supported by the Shanxi Province Grant(202203021212007,2023SHB003).
文摘Hydrogen peroxide(H_(2)O_(2))oxidation and reduction reactions(HPOR/HPRR)are pivotal in various innovative electrochemical energy conversion devices.A comprehensive understanding of these mechanisms is critical for catalyst design and performance improvement in these applications.In this work,we systematically investigate the HPOR/HPRR mechanisms on low-index Pt surfaces,specifically Pt(111),Pt(100)and Pt(110),through density functional theory(DFT)calculations combined with the computational hydrogen electrode(CHE)model.For HPOR,all the low-index Pt surfaces exhibit a unified potential-determining step(PDS)involving the electrochemical oxidation of hydroperoxyl intermediates(HOO*).The binding free energy of HOO*(Δ_(GHOO*))emerges as an activity descriptor,with Pt(110)exhibiting the highest HPOR activity.The HPRR mechanism follows a chem-electrochemical(C-EC)pathway.The rate-determining step(RDS)of HPRR is either the cleavage of the HO-OH bond(chemical)or the reduction of HO(electrochemical),depending on their respective activation energies.These activation energies are functions of the HO*binding free energy,Δ_(GHO*),establishingΔ_(GHO*)as the descriptor for HPRR activity prediction.Pt(111)and Pt(100)are identified as the most active HPRR catalysts among the studied metal surfaces,although they still experience a significant overpotential.The scaling relationship betweenΔ_(GHOO*)andΔ_(GHO*)reveals a thermodynamic coupling of HPOR and HPRR,explaining their occurrence on Pt surfaces.These findings provide important insights and activity descriptors for both HPOR and HPRR,providing valuable guidance for the design of electrocatalysts in H_(2)O_(2)-related energy applications and fuel cells.
文摘Underwater pipeline inspection plays a vital role in the proactive maintenance and management of critical marine infrastructure and subaquatic systems.However,the inspection of underwater pipelines presents a challenge due to factors such as light scattering,absorption,restricted visibility,and ambient noise.The advancement of deep learning has introduced powerful techniques for processing large amounts of unstructured and imperfect data collected from underwater environments.This study evaluated the efficacy of the You Only Look Once(YOLO)algorithm,a real-time object detection and localization model based on convolutional neural networks,in identifying and classifying various types of pipeline defects in underwater settings.YOLOv8,the latest evolution in the YOLO family,integrates advanced capabilities,such as anchor-free detection,a cross-stage partial network backbone for efficient feature extraction,and a feature pyramid network+path aggregation network neck for robust multi-scale object detection,which make it particularly well-suited for complex underwater environments.Due to the lack of suitable open-access datasets for underwater pipeline defects,a custom dataset was captured using a remotely operated vehicle in a controlled environment.This application has the following assets available for use.Extensive experimentation demonstrated that YOLOv8 X-Large consistently outperformed other models in terms of pipe defect detection and classification and achieved a strong balance between precision and recall in identifying pipeline cracks,rust,corners,defective welds,flanges,tapes,and holes.This research establishes the baseline performance of YOLOv8 for underwater defect detection and showcases its potential to enhance the reliability and efficiency of pipeline inspection tasks in challenging underwater environments.
基金partially supported by the Natural Science Foundation of Shandong Province,China(No.ZR2023ME009)the National Natural Science Foundation of China(No.51909252)。
文摘Accurately estimating depth from underwater monocular images is essential for the target tracking task of unmanned underwater vehicles.This work proposes a method based on the Lpg-Lap Unet architecture.First,the Unet architecture integrates Laplacian pyramid depth residuals and Sobel operators to improve the boundary details in depth images,which may suffer from the feature loss caused by upsampling and the blurriness of underwater images.Multiscale local planar guidance layers then fully exploit the intermediate depth features,and a comprehensive loss function ensures robustness and accuracy.Experimental results on benchmarks demonstrate the effectiveness of Lpg-Lap Unet and its superior performance over state-of-the-art models.An underwater target tracking system is then designed to further validate its real-time capabilities in the AirSim simulation platform.
基金the Scientific and Technological Research Council of Türkiye(TÜBiTAK).
文摘This study investigated surface roughness,the wettability behavior,and surface energy of Co-based alloy specimens textured using the biomimetic Laser Surface Texturing(LST)method.The surface texture was inspired by the patterns found on marine shells.The impacts of the parameters on wettability,Surface Free Energy(SFE),surface topography,and texture roughness generated by the laser beam tracking a spiral path were investigated.Reducing spiral pitch produces more complicated and chaotic surface patterns.Most surfaces are hydrophobic,and surface roughness and topography influence the Contact Angle(CA).Topography and roughness were affected by frequency and scanning speed;a decrease in scanning speed and frequency generated more chaotic and irregular surface textures.With general factorial analysis and Analysis of Variance(ANOVA),our statistical study reveals that accounting for 88%of the influence,the scanning speed is the primary factor influencing surface roughness.On the other hand,the spiral pitch is essential for defining the struc-tural features of the surface,even if it less influences roughness.The SFE of laser-textured CoCr28Mo alloy specimens was optimizable within the range of 14-32 mN/m.The relevant findings offer valuable insights into optimizing LST for the specific surface properties of the Co-based alloy.
基金supported by the Science and Technology Research Project of Henan Province(No.232102241014)the Key scientific research project of Henan Province Colleges and Universities(No.22A470002)Doctoral Fund Project of Henan Polytechnic University(No.B2021-37).
文摘This study experimentally investigates the influence of surfacewettability on the frosting characteristics of three types of corrugated structures(Types A,B,and C)under controlled low-temperature conditions.The experiments were conducted in a constant-temperature bath at a cold surface temperature of–5℃,relative humidity of 90%,and ambient air temperature of 10℃.The results reveal that the variation trends of frost morphology,frost mass,and frost layer thickness are generally consistent across surfaces with different wettability.Among the tested surfaces,frost crystal formation and complete surface coverage occurred latest on the superhydrophobic surface(CA=153.9–165.8℃),next on the bare aluminumsurface(75.3–83.2℃),and earliest on the hydrophilic surface(5.3–7.5℃).At the same frosting duration,the superhydrophobic surface exhibited a sparse and fluffy frost layer,the bare aluminum surface formed a rough and dense frost,while the hydrophilic surface developed a fine and compact frost layer.The amount of frost formation decreased in the order of hydrophilic>bare aluminum>superhydrophobic,indicating that the superhydrophobic surface provides the most significant anti-frosting effect during the initial stages of frost formation.For instance,on the Type A corrugated structure,after 15 min of frosting,the frost mass on the superhydrophobic surface was 38.78%and 68.45%lower than those on the bare aluminum and hydrophilic surfaces,respectively.After 30 min,these differences were 4.99%and 25.26%,respectively.Overall,the superhydrophobic surface exhibited the smallest frost mass and frost layer thickness,demonstrating superior anti-frosting performance compared with the other surface types.
基金The National Natural Science Foundation of China(Grant No.52201376)the Natural Science Foundation of Hubei Province,China(Grant No.2023AFB683).
文摘In this investigation,a hybrid approach integrating the IDDES turbulence model and FW-H is employed to forecast the hydroacoustic of the rim driven thruster(RDT)under non-cavitation and uniform flow conditions at varying loading conditions(J=0.3 and J=0.6).It is revealed that the quadrupole term contribution in the P-FWH method significantly affects the monopole term in the low-frequency region,while it mainly affects the dipole term in the high-frequency region.Specifically,the overall sound pressure levels(SPL)of the RDT using the P-FWH method are 2.27 dB,10.03 dB,and 16.73 dB at the receiving points from R1 to R3 under the heavy-loaded condition,while they increase by 0.67 dB at R1,and decrease by 14.93 dB at R2,and 22.20 dB at R3,for the light-loaded condition.The study also utilizes the pressure-time derivatives to visualize the numerical noise and to pinpoint the dynamics of the vortex cores,and the optimization of the grid design can significantly reduce the numerical noise.The computational accuracy of the P-FWH method can meet the noise requirements for the preliminary design of rim driven thrusters.
基金Supported by the National Natural Science Foundation of China(No.51905211)A Project of the“20 Regulations for New Universities”Funding Program of Jinan(No.202228116).
文摘Vortex-induced vibration(VIV)of an underwater manipulator in pulsating flow presents a notable engineering problem in precise control due to the velocity variation in the flow.This study investigates the VIV response of an underwater manipulator subjected to pulsating flow,focusing on how different postures affect the behavior of the system.The effects of pulsating parameters and manipulator arrangement on the hydrodynamic coefficient,vibration response,motion trajectory,and vortex shedding behaviors were analyzed.Results indicated that the cross flow vibration displacement in pulsating flow increased by 32.14%compared to uniform flow,inducing a shift in the motion trajectory from a crescent shape to a sideward vase shape.In the absence of interference between the upper and lower arms,the lift coefficient of the manipulator substantially increased with rising pulsating frequency,reaching a maximum increment of 67.0%.This increase in the lift coefficient led to a 67.05%rise in the vibration frequency of the manipulator in the in-line direction.As the pulsating amplitude increased,the drag coefficient of the underwater manipulator rose by 36.79%,but the vibration frequency in the cross-flow direction decreased by 56.26%.Additionally,when the upper and lower arms remained in a state of mutual interference,the cross-flow vibration amplitudes of the upper and lower arms were approximately 1.84 and 4.82 times higher in a circular-elliptical arrangement compared to an elliptical-circular arrangement,respectively.Consequently,the flow field shifted from a P+S pattern to a disordered pattern,disrupting the regularity of the motion trajectory.
基金supported by the National Natural Science Foundation of China (52172228)the Natural Science Foundation of Fujian Province (2024J01475 and 2023J05127)
文摘Lithium-sulfur batteries(LSBs)represent a next-generation energy storage technology,but widespread applications are restricted by the shuttle of lithium polysulfides(LiPSs).The rational design of separators has been demonstrated to be one of the most efficient and cost-effective strategies to curb the shuttle effect,and tremendous research progress has been achieved.The efficiency of a separator depends on its interaction with LiPSs,which is governed by the surface energy and binding strength.Despite several review works that have been reported to advance the separators,most of them primarily focus on active material innovation and construction.The most crucial issues of surface binding energy have not been systematically reviewed,limiting the precise design of efficient separators.In this review,fundamentals related to surface energy and binding interactions with LiPSs are comprehensively analyzed and discussed.With surface binding and energy main lines,the advancements in separator engineering strategies are elaborately summarized and discussed.Moreover,techniques for evaluating affinity to LiPSs are thoroughly analyzed to avoid any ambiguities in measurement.Based on the research context,valuable research directions are suggested to construct efficient separators.This work provides guidelines to regulate the surface binding and energy of separators for high-performance LSBs.
基金financial support from the National Natural Science Foundation of China(No.52035012)the Guangdong Basic and Applied Basic Research Foundation(No.2025A1515012203)。
文摘The specific surface area(S S)and pore size(D)exhibit an inherent trade-off in the microscale design of bone implants:larger pores typically correlate with reduced surface area and vice versa.This relationship has attracted notable attention because of its critical role in the regulation of cell adhesion and osteogenesis.However,it remains largely unclear how S S and D affect the generated bone tissue and dynamically change during long-term osteogenesis.Herein,by applying rigorous geometric mapping to minimal surfaces,we constructed precisely partitioned and layer-by-layer thickened tissue models to simulate osteogenesis across different temporal scales and thereby track the dynamic evolution of geometric characteristics,permeability,and mechanobiological tissue differentiation.The high-S S samples were found to facilitate the rapid formation of new bone tissue in the early stages.However,their smaller pores tended to cause occlusions,hindering further tissue development.In contrast,low-S S samples showed slower bone regeneration,but their larger pores provided adequate physical space for tissue regeneration and mass transport,ultimately promoting bone formation in the long term.Mechanobiological regulation suggests that fibrous tissue formation inhibits additional bone formation,establishing a dynamic equilibrium between osteogenesis and pore space to sustain nutrient/waste exchange throughout the regenerative process.Overall,smaller pores are preferable in implants for minimally loaded osteoplasty procedures focused on early-stage bone consolidation,whereas larger pores are preferable in dynamically loaded implants requiring prolonged mechanical stability.
文摘The original online version of this article was revised:Several errors occurred in the published version of the article.These have now been corrected as follows:Page 2,section"2.2 Laser Texturing Procedure of Surfaces",line 2:The device name was corrected from"YDFLP-E-50-M8"to"YDFLP-50-M8."Page 3,Section 2.4:The phrase"95%confidence interval"has been corrected to"95%confidence level."Page 3,Figure 1 caption:The phrase"fandg"has been corrected to"f and g."The order"C4 and C12"has been reversed to"C12 and C4,"in accordance with the display order in the figure.Page 4,Figure reference:The phrase"Figs.4c and d"has been corrected to"Figs.5b and c."Page 5,paragraph starting with"The ANOVA results are presented...":The phrase"95%confidence interval"has been corrected to"95%confidence level."
文摘Photodetectors can convert light energy into electrical signals,so are widely used in photovoltaics,photon counting,monitoring,and imaging.Photodetectors are easy to prepare high-resolution photochips because of their small size unit integration.However,these photodetector units often exhibit poor photoelectric performance due to material defects and inadequate structures,which greatly limit the functions of devices.Designing modification strategies and micro-/nanostructures can compensate for defects,adjust the bandgap,and develop novel quantum structures,which consequently optimize photovoltaic units and revolutionize optoelectronic devices.Here,this paper aims to comprehensively elaborate on the surface/interface engineering scheme of micro-/nano-photodetectors.It starts from the fundamentals of photodetectors,such as principles,types,and parameters,and describes the influence of material selection,manufacturing techniques,and post-processing.Then,we analyse in detail the great influence of surface/interface engineering on the performance of photovoltaic devices,including surface/interface modification and micro-/nanostructural design.Finally,the applications and prospects of optoelectronic devices in various fields such as miniaturization of electronic devices,robotics,and human–computer interaction are shown.
基金supported in part by the National Natural Science Foundation of China(Key Program)under Grant No.62031021。
文摘Cascading failures pose a serious threat to the survivability of underwater unmanned swarm networks(UUSNs),significantly limiting their service ability in collaborative missions such as military reconnaissance and environmental monitoring.Existing failure models primarily focus on power grids and traffic systems,and don't address the unique challenges of weak-communication UUSNs.In UUSNs,cascading failure present a complex and dynamic process driven by the coupling of unstable acoustic channels,passive node drift,adversarial attacks,and network heterogeneity.To address these challenges,a directed weighted graph model of UUSNs is first developed,in which node positions are updated according to ocean-current-driven drift and link weights reflect the probability of successful acoustic transmission.Building on this UUSNs graph model,a cascading failure model is proposed that integrates a normal-failure-recovery state-cycle mechanism,multiple attack strategies,and routingbased load redistribution.Finally,under a five-level connectivity UUSNs scheme,simulations are conducted to analyze how dynamic topology,network load,node recovery delay,and attack modes jointly affect network survivability.The main findings are:(1)moderate node drift can improve survivability by activating weak links;(2)based-energy routing(BER)outperform based-depth routing(BDR)in harsh conditions;(3)node self-recovery time is critical to network survivability;(4)traditional degree-based critical node metrics are inadequate for weak-communication UUSNs.These results provide a theoretical foundation for designing robust survivability mechanisms in weak-communication UUSNs.
基金Wuxi University Research Start-up Fund for Introduced Talents(Grant No:2024r031)Technology Development Contract(Contract Registration Number:2024320205000963)+1 种基金National Natural Science Foundation of China(Grant No.52275288)Ningbo Key Research and Development Plan(Grant No.2023Z022).
文摘To solve the problem of abnormal abrasion of Cu-Based Friction Materials(CBFMs),Bionic Non-Smooth Surface(BNS)on friction surface of CBFMs was constructed based on bionic principles,and the optimal bionic prototype was selected by Finite Element Method(FEM).In addition,the bionic parameters were optimized by Response Surface Method(RSM).Samples holding BNS were prepared by Laser Processing,tribological properties were tested by a Friction and Wear Tester and worn surface morphology was characterized by a Scanning Electron Microscope(SEM).The results showed that BNS on friction surface could regulate the stress distribution and alleviate the peak stress.Among all samples,the coupled texture of pit-hexagonal got the minimum peak stress.During braking,bionic texture could also collect wear debris or change the motion forms from sliding to rotation,which can reduce abnormal abrasion.The wear rate was reduced by 19.31%.The results in this paper can provide a new idea for enhancing the tribological properties of CBFMs,and can also lay the foundation for further research of bionic tribology.
文摘In this study,polyacrylic acid(PAA)films were employed as a model system,and a series of PAA films with tunable water wettability was systematically prepared by varying molecular weight and curing temperature.Using attenuated total reflectance Fourier-transform infrared spectroscopy(ATR-FTIR),the molecular configurations of surface carboxyl groups(COOH),free carboxyl(COOH_(f))and hydrogen-bonded carboxyl(COOH_(HB),were directly correlated with the polar component of surface energy(γ^(s,p)).By decomposing theγ^(s,p)values of the PAA thin films as a sum of the contributions of COOH_(f)and COOH_(H B),the intrinsic polar component of surface energy of COOH_(H B)(γ_(H B)^(s,p*))was quantified for the first time as 8.34 mN/m,significantly lower than that of COOH_(f)(γ_(f)^(s,p*)=34 mN/m).This result highlights that hydrogen bonding markedly reduces theγ^(s,p),providing a rational explanation for the relatively large water contact angle observed on PAA thin films.Furthermore,it establishes a thermodynamic basis for estimating the fraction of surface COOH_(H B)groups(f H B)from wettability measurements.Further extension of the model to carboxyl-terminated self-assembled monolayers(COOH-SAMs)revealed that surface COOH density(ΣCOOH)critically regulates wetting behavior:whenΣCOOH ranges from 4.30 to 5.25 nm^(-2),COOH groups predominantly exist in a free state and facilitate effective hydration layers,thereby promoting superhydrophilicity.Overall,this study not only establishes a unified thermodynamic framework linking surface COOH configurations to macroscopic wettability,but also validates its universality by extending it to COOH-SAMs systems,thereby providing a unified theoretical framework for the controllable design of hydrophilicity in various COOH-functionalized surfaces.
基金supported by the National Key Research and Development Program of China(2023YFB3907304-3)the National Natural Science Foundation of China(NSFC)(62271050)。
文摘A methodology for the reduction of radar cross section(RCS)of cambered platforms within the target airspace is presented,which utilizes a dual-polarized ultra-wide-angle artificial electromagnetic absorbing surface.By applying the theory of generalized Brewster complex wave impedance matching,five distinct unit cell designs are developed to attain more than95%absorption rate for dual-polarized incident waves within five angular ranges:0°-30°,30°-50°,50°-60°,60°-70°,and 70°-80°.To optimally reduce the RCS of a cambered platform,the five types of units can be evenly distributed on the surface based on the local incident angles of plane waves originating from the target airspace.As an illustrative example,the leading edge of an airfoil is taken into account,and experimental measurements validate the efficiency of the proposed structure.Specifically,the absorbing surface achieves more than 10 dB of RCS reduction in the frequency ranges from 5-10 GHz(about66.7%relative bandwidth)for dual polarizations.
基金supported in part by the National Natural Science Foundation of China(Grant No.62071123)in part by the Natural Science Foundation of Fujian Province(Grant Nos.2024J01971,2022J05202)in part by the Young and Middle-Aged Teacher Education Research Project of Fujian Province(Grant No.JAT210370).
文摘In industrial manufacturing,efficient surface defect detection is crucial for ensuring product quality and production safety.Traditional inspectionmethods are often slow,subjective,and prone to errors,while classicalmachine vision techniques strugglewith complex backgrounds and small defects.To address these challenges,this study proposes an improved YOLOv11 model for detecting defects on hot-rolled steel strips using the NEU-DET dataset.Three key improvements are introduced in the proposed model.First,a lightweight Guided Attention Feature Module(GAFM)is incorporated to enhance multi-scale feature fusion,allowing the model to better capture and integrate semantic and spatial information across different layers,which improves its ability to detect defects of varying sizes.Second,an Aggregated Attention(AA)mechanism is employed to strengthen the representation of critical defect features while effectively suppressing irrelevant background information,particularly enhancing the detection of small,low-contrast,or complex defects.Third,Ghost Dynamic Convolution(GDC)is applied to reduce computational cost by generating low-cost ghost features and dynamically reweighting convolutional kernels,enabling faster inference without sacrificing feature quality or detection accuracy.Extensive experiments demonstrate that the proposed model achieves a mean Average Precision(mAP)of 87.2%,compared to 81.5%for the baseline,while lowering computational cost from6.3Giga Floating-point Operations Per Second(GFLOPs)to 5.1 GFLOPs.These results indicate that the improved YOLOv11 is both accurate and computationally efficient,making it suitable for real-time industrial surface defect detection and contributing to the development of practical,high-performance inspection systems.
