Arrhythmias are a frequently occurring phenomenon in clinical practice,but how to accurately dis-tinguish subtle rhythm abnormalities remains an ongoing difficulty faced by the entire research community when conductin...Arrhythmias are a frequently occurring phenomenon in clinical practice,but how to accurately dis-tinguish subtle rhythm abnormalities remains an ongoing difficulty faced by the entire research community when conducting ECG-based studies.From a review of existing studies,two main factors appear to contribute to this problem:the uneven distribution of arrhythmia classes and the limited expressiveness of features learned by current models.To overcome these limitations,this study proposes a dual-path multimodal framework,termed DM-EHC(Dual-Path Multimodal ECG Heartbeat Classifier),for ECG-based heartbeat classification.The proposed framework links 1D ECG temporal features with 2D time–frequency features.By setting up the dual paths described above,the model can process more dimensions of feature information.The MIT-BIH arrhythmia database was selected as the baseline dataset for the experiments.Experimental results show that the proposed method outperforms single modalities and performs better for certain specific types of arrhythmias.The model achieved mean precision,recall,and F1 score of 95.14%,92.26%,and 93.65%,respectively.These results indicate that the framework is robust and has potential value in automated arrhythmia classification.展开更多
Understanding how rock slopes respond to blasting loads is crucial for maintaining excavation safety and slope stability.Nevertheless,the spatiotemporal evolution,nonlinear dependence on blasting parameters,and predic...Understanding how rock slopes respond to blasting loads is crucial for maintaining excavation safety and slope stability.Nevertheless,the spatiotemporal evolution,nonlinear dependence on blasting parameters,and predictive behavior of dominant frequency responses in slope vibrations remain insufficiently understood and quantified.This study combines time-frequency analysis with machine learning to explore how the dominant frequency(f_(d))evolves in slopes under blasting.Continuous Wavelet Transform(CWT)was employed to characterize the temporal-frequency evolution of vibration signals,revealing that the dominant frequency exhibits strong spatial dependence and nonlinear variability influenced by blasting parameters and rock mass structures.Three machine learning models,namely Back Propagation Neural Network(BP),Support Vector Machine(SVM),and Random Forest(RF),were developed to predict f_(d) based on 1,000 monitoring samples obtained from numerical and field simulations.Among them,the RF model achieved the highest prediction accuracy,with mean absolute percentage errors(MAPE)below 15%,demonstrating strong robustness and generalization capability.Our analysis shows that external excitation factors,especially the loading frequency(f_(d)),mainly control the frequency response,while internal controlling factors,such as spatial position,lithological variation,and mechanical heterogeneity,modulate localized frequency amplification and energy redistribution.The results reveal that f_(d) tends to decrease with elevation and distance from the blasting source,whereas structural planes and weathered zones induce high-frequency amplification due to scattering and modal coupling effects.This study offers a new framework combining time-frequency analysis and machine learning to measure the nonlinear interaction between blasting and rock mass response,offering new insights for dynamic stability evaluation and hazard mitigation in complex rock slope systems.展开更多
The state-of-the-art optical atomic clocks and the time-frequency signal transmission open a fresh field for gravity potential(geopotential)determination.Various methods,including optical fiber frequency transfer,sate...The state-of-the-art optical atomic clocks and the time-frequency signal transmission open a fresh field for gravity potential(geopotential)determination.Various methods,including optical fiber frequency transfer,satellite two-way,satellite common-view,satellite carrier phase,VLBI,tri-frequency combination,and dual-frequency combination,were developed to determine the geopotential differences using optical atomic clocks and then determine the geopotential at station B based on the geopotential at station A.This review elaborates the principles,methods,scientific objectives,applications,and relevant research trends of geopotential determination based on time-frequency signals.展开更多
A dynamic graph(DG)is adopted to portray the evolving interplay between nodes in real-world scenarios prevalently.A high-order graph convolutional network(HGCN)is equipped with the ability to represent a DG by the spa...A dynamic graph(DG)is adopted to portray the evolving interplay between nodes in real-world scenarios prevalently.A high-order graph convolutional network(HGCN)is equipped with the ability to represent a DG by the spatial-temporal message passing mechanism built on tensor product.Concretely,an HGCN utilizes the discrete Fourier transform(DFT)to implement temporal message passing and then employs face-wise product to realize spatial message passing.However,DFT is only a special case of assorted time-frequency transforms,which considers the complex temporal patterns partially,thereby resulting in an inaccurate temporal message passing possibly.To address this issue,this study proposes six advanced time-frequency transform-incorporated HGCNs(TF-HGCNs)with discrete Fourier,discrete Hartley,discrete cosine,Haar wavelet,Walsh Hadamard,and slant transforms.In addition,a potent ensemble is built regarding the proposed six TF-HGCNs as the bases.Finally,the corresponding theoretical proof is presented.Empirical studies on six DG datasets demonstrate that owing to diverse time-frequency transforms,the proposed six TF-HGCNs significantly outperform state-of-the-art models in addressing the task of link weight estimation.Moreover,their ensemble outstrips each base's performance.展开更多
Quantitative detection of sleeve grouting compactness is a technical challenge in civil engineering testing.This study explores a novel quantitative detection method based on ultrasonic time-frequency dual-domain anal...Quantitative detection of sleeve grouting compactness is a technical challenge in civil engineering testing.This study explores a novel quantitative detection method based on ultrasonic time-frequency dual-domain analysis.It establishes a mapping relationship between sleeve grouting compactness and characteristic parameters.First,this study made samples with gradient defects for two types of grouting sleeves,G18 and G20.These included four cases:2D,4D,6D defects(where D is the diameter of the grouting sleeve),and no-defect.Then,an ultrasonic input/output data acquisition system was established.Three-dimensional sound field distribution data were obtained through an orthogonal detection layout and pulse reflection principles.Finally,a novel quantification detection with a comprehensive defect index(DI)was established by comprehensively considering eight feature parameters,such as time-frequency domain Kurtosis factor(KU),Skewness factor(SK),Formfactor(FF),Crest factor(CF),Impulse factor(IF),Clearance factor(CLF),Wavelet packet energy entropy(WPEE),and Hilbert energy peak(HEP).Construct a DI index by quantifying the difference between defect signals and defect free signals in the time-frequency domain.Experimental results show that,under no-defect conditions,the values of feature parameters are significantly lower than those under defect conditions.Among these,the KU,FF,CF,WPEE and HEP exhibit strong correlations with grout sleeve compactness.The proposed DI index in both types of grout sleeves showed good universality with a linear fit goodness of 0.847–0.962.However,G20 the larger inner diameter and length of the sleeve result in a more complex medium effect during ultrasonic propagation,making its DI index more sensitive to defects than the G18 sleeve.Therefore,the presented method is effective for quantitative detection and analysis of the compactness of grouting sleeves.展开更多
Fast wavelet multi-resolution analysis (wavelet MRA) provides a effective tool for analyzing and canceling disturbing components in original signal. Because of its exponential frequency axis, this method isn't s...Fast wavelet multi-resolution analysis (wavelet MRA) provides a effective tool for analyzing and canceling disturbing components in original signal. Because of its exponential frequency axis, this method isn't suitable for extracting harmonic components. The modified exponential time-frequency distribution ( MED) overcomes the problems of Wigner distribution( WD) ,can suppress cross-terms and cancel noise further more. MED provides high resolution in both time and frequency domains, so it can make out weak period impulse components fmm signal with mighty harmonic components. According to the 'time' behavior, together with 'frequency' behavior in one figure,the essential structure of a signal is revealed clearly. According to the analysis of algorithm and fault diagnosis example, the joint of wavelet MRA and MED is a powerful tool for fault diagnosis.展开更多
A critical problem facing data collection in structural health monitoring,for instance via sensor networks,is how to extract the main components and useful features for damage detection.A structural dynamic measuremen...A critical problem facing data collection in structural health monitoring,for instance via sensor networks,is how to extract the main components and useful features for damage detection.A structural dynamic measurement is more often a complex time-varying process and therefore,is prone to dynamic changes in time-frequency contents.To extract the signal components and capture the useful features associated with damage from such nonstationary signals,a technique that combines the time and frequency analysis and shows the signal evolution in both time and frequency is required.Wavelet analyses have proven to be a viable and effective tool in this regard.Wavelet transform(WT)can analyze different signal components and then comparing the characteristics of each signal with a resolution matched to its scale.However,the challenge is the selection of a proper wavelet since various wavelets with varied properties that are to analyze the same data may result in different results.This article presents a study on how to carry out a comparative analysis based on analytic wavelet scalograms,using structural dynamic acceleration responses,to evaluate the effectiveness of various wavelets for damage detection in civil structures.The scalogram’s informative time-frequency regions are examined to analyze the variation of wavelet coefficients and show how the frequency content of a signal changes over time to detect transient events due to damage.Subsequently,damage-induced changes are tracked with time-frequency representations.Towards this aim,energy distribution and sharing information are investigated.The undamaged and damaged simulated comparative results of a structure reveal that the damaged structure were shifted from the undamaged structure.Also,the Bump wavelet shows the best results than the others.展开更多
In the time-frequency analysis of seismic signals, the matching pursuit algorithm is an effective tool for non-stationary signals, and has high time-frequency resolution and a transient structure with local self-adapt...In the time-frequency analysis of seismic signals, the matching pursuit algorithm is an effective tool for non-stationary signals, and has high time-frequency resolution and a transient structure with local self-adaption. We expand the time-frequency dictionary library with Ricker, Morlet, and mixed phase seismic wavelets, to make the method more suitable for seismic signal time-frequency decomposition. In this paper, we demonstrated the algorithm theory using synthetic seismic data, and tested the method using synthetic data with 25% noise. We compared the matching pursuit results of the time-frequency dictionaries. The results indicated that the dictionary which matched the signal characteristics better would obtain better results, and can reflect the information of seismic data effectively.展开更多
Vibration signals have the characteristics of multi-source strong shock coupling and strong noise interference owing to the complex structure of reciprocating machinery.Therefore,it is difficult to extract,analyze,and...Vibration signals have the characteristics of multi-source strong shock coupling and strong noise interference owing to the complex structure of reciprocating machinery.Therefore,it is difficult to extract,analyze,and diagnose mechanical fault features.To accurately extract sensitive features from the strong noise interference and unsteady monitoring signals of reciprocating machinery,a study on the time-frequency feature extraction method of multi-source shock signals is conducted.Combining the characteristics of reciprocating mechanical vibration signals,a targeted optimization method considering the variational modal decomposition(VMD)mode number and second penalty factor is proposed,which completed the adaptive decomposition of coupled signals.Aiming at the bilateral asymmetric attenuation characteristics of reciprocating mechanical shock signals,a new bilateral adaptive Laplace wavelet(BALW)is established.A search strategy for wavelet local parameters of multi-shock signals is proposed using the harmony search(HS)method.A multi-source shock simulation signal is established,and actual data on the valve fault are obtained through diesel engine fault experiments.The fault recognition rate of the intake and exhaust valve clearance is above 90%and the extraction accuracy of the shock start position is improved by 10°.展开更多
An air gun generates acoustic signals for seismic exploration by releasing a high-pressure gas.A large error is always gradually introduced into the ideal-gas model when the pressure in the air-gun chamber exceeds 100...An air gun generates acoustic signals for seismic exploration by releasing a high-pressure gas.A large error is always gradually introduced into the ideal-gas model when the pressure in the air-gun chamber exceeds 100 atm.In the van der Waals non-ideal-gas theory,the gas in the air gun can be regarded as an actual gas,and the error is less than 2%.The van der Waals model is established in combination with the quasi-static open thermodynamic system and bubble-motion equation by considering the bubble rise,bubble interaction,and throttling eff ect.The mismatch between the van der Waals and ideal-gas models is related to the pressure.Theoretically,under high-pressure conditions,the van der Waals air-gun model yields results that are closer to the measured results.Marine vertical cables are extended to the seafl oor using steel cables that connect the cement blocks,but the corresponding hydrophones are suspended in the seawater.Thus,noise associated with ships,ocean surges,and coupling problems is avoided,and the signal-to-noise ratio and resolution of marine seismic data are improved.This acquisition method satisfies the conditions of recording air-gun far-fi eld wavelets.According to an actual vertical-cable observation system,the van der Waals air-gun model is used to model the wavelet of different azimuth and take-off angles.The characteristics of the experimental and simulated data demonstrate good agreement,which indicates that the van der Waals method is accurate and reliable.The accuracy of the model is directly related to the resolution,thus aff ecting the resolution ability of the stratum.展开更多
Through analyzing the near-field hydrophone records of the airgun experiment in the Jiemian reservoir,Fujian,we study the time-frequency characteristic of airgun source wavelet and the influence of gun depth and firin...Through analyzing the near-field hydrophone records of the airgun experiment in the Jiemian reservoir,Fujian,we study the time-frequency characteristic of airgun source wavelet and the influence of gun depth and firing pressure,and explain the process of bubble oscillation based on the Johnson( 1994) bubble model. The data analysis shows that:( 1) Airgun wavelet is composed of primary pulse and bubble pulse. The primary pulse,which is of large amplitude,short duration and wide frequency band,is usually used in shallow exploration. The bubble pulse,which is concentrated in the low-frequency range,is usually used in deep exploration with deep vertical penetration and far horizontal propagation.( 2) The variation of primary pulse amplitude with gun depth is very small,bubble pulse amplitude and the dominant frequency increase,and peak-bubble ratio and bubble period decrease. When the gun depth is 10 m,primary pulse amplitude and peakbubble ratio are maximum,which is suitable for shallow exploration; when gun depth is25 m,bubble pulse amplitude is large, and peak-bubble ratio is minimum, which is suitable for deep exploration.( 3) The primary pulse amplitude,bubble pulse amplitude,peak-bubble ratio,and bubble period increase and the dominant frequency decreases with increased firing pressure.展开更多
With the increasingly complex and changeable electromagnetic environment,wireless communication systems are facing jamming and abnormal signal injection,which significantly affects the normal operation of a communicat...With the increasingly complex and changeable electromagnetic environment,wireless communication systems are facing jamming and abnormal signal injection,which significantly affects the normal operation of a communication system.In particular,the abnormal signals may emulate the normal signals,which makes it very challenging for abnormal signal recognition.In this paper,we propose a new abnormal signal recognition scheme,which combines time-frequency analysis with deep learning to effectively identify synthetic abnormal communication signals.Firstly,we emulate synthetic abnormal communication signals including seven jamming patterns.Then,we model an abnormal communication signals recognition system based on the communication protocol between the transmitter and the receiver.To improve the performance,we convert the original signal into the time-frequency spectrogram to develop an image classification algorithm.Simulation results demonstrate that the proposed method can effectively recognize the abnormal signals under various parameter configurations,even under low signal-to-noise ratio(SNR)and low jamming-to-signal ratio(JSR)conditions.展开更多
The local wave method is a very good time-frequency method for nonstationaryvibration signal analysis. But the interfering noise has a big influence on the accuracy oftime-frequency analysis. The wavelet packet de-noi...The local wave method is a very good time-frequency method for nonstationaryvibration signal analysis. But the interfering noise has a big influence on the accuracy oftime-frequency analysis. The wavelet packet de-noising method can eliminate the interference ofnoise and improve the signal-noise-ratio. This paper uses the local wave method to decompose thede-noising signal and perform a time-frequency analysis. We can get better characteristics. Finally,an example of wavelet packet de-noising and a local wave time-frequency spectrum application ofdiesel engine surface vibration signal is put forward.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
Image captioning,a pivotal research area at the intersection of image understanding,artificial intelligence,and linguistics,aims to generate natural language descriptions for images.This paper proposes an efficient im...Image captioning,a pivotal research area at the intersection of image understanding,artificial intelligence,and linguistics,aims to generate natural language descriptions for images.This paper proposes an efficient image captioning model named Mob-IMWTC,which integrates improved wavelet convolution(IMWTC)with an enhanced MobileNet V3 architecture.The enhanced MobileNet V3 integrates a transformer encoder as its encoding module and a transformer decoder as its decoding module.This innovative neural network significantly reduces the memory space required and model training time,while maintaining a high level of accuracy in generating image descriptions.IMWTC facilitates large receptive fields without significantly increasing the number of parameters or computational overhead.The improvedMobileNet V3 model has its classifier removed,and simultaneously,it employs IMWTC layers to replace the original convolutional layers.This makes Mob-IMWTC exceptionally well-suited for deployment on lowresource devices.Experimental results,based on objective evaluation metrics such as BLEU,ROUGE,CIDEr,METEOR,and SPICE,demonstrate that Mob-IMWTC outperforms state-of-the-art models,including three CNN architectures(CNN-LSTM,CNN-Att-LSTM,CNN-Tran),two mainstream methods(LCM-Captioner,ClipCap),and our previous work(Mob-Tran).Subjective evaluations further validate the model’s superiority in terms of grammaticality,adequacy,logic,readability,and humanness.Mob-IMWTC offers a lightweight yet effective solution for image captioning,making it suitable for deployment on resource-constrained devices.展开更多
Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional ...Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at https://github.com/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).展开更多
In wave-equation migration and demigration,the cross-correlation imaging/forwarding step implicitly injects an additional copy of the source wavelet,so that the amplitude spectrum of the wavelet is applied redundantly...In wave-equation migration and demigration,the cross-correlation imaging/forwarding step implicitly injects an additional copy of the source wavelet,so that the amplitude spectrum of the wavelet is applied redundantly(effectively imposing a wavelet-spectrum weighting,often akin to an amplitude-squared bias).This redundancy degrades structural fidelity and amplitude balance yet is frequently overlooked.We(i)formalize the mechanism by which cross-correlation duplicates the source-wavelet amplitude effect in both migration and demigration,and(ii)introduce a source-equalized operator that removes the redundancy by deconvolving(or dividing by)the wavelet amplitude spectrum in the imaging condition and its demigration counterpart,while leaving phase/kinematics intact.Using a band-limited Ricker wavelet on a two-layer model and on Marmousi,we show that,if unmanaged,the redundant wavelet spectrum broadens main lobes,introduces ringing,and suppresses vertical resolution in migrated images,and inflates spectrum mismatches between demigrated and observed data even when peak times agree.With our correction,images recover observed-data-consistent bandwidth and sharpened interfaces,and demigrated data also exhibit improved spectrum conformity and reduced amplitude misfit.The results clarify when source amplitudes matter,why cross-correlation makes them redundantly matter,and how a lightweight spectral correction restores physically meaningful amplitude behavior in wave-equation migration/demigration.展开更多
Scientific analysis of aeolian sand environments is fundamental for sustainable disaster mitigation along desert highways.However,significant regional variability in wind energy conditions complicates accurate charact...Scientific analysis of aeolian sand environments is fundamental for sustainable disaster mitigation along desert highways.However,significant regional variability in wind energy conditions complicates accurate characterization of wind regimes and introduces uncertainty in determining optimal monitoring timescales.Moreover,prevailing sand control measures often rely on standardized designs rather than site-specific adaptive strategies.To address these issues,this study proposes an integrated framework for aeolian environment analysis and develops targeted disaster mitigation strategies tailored for desert highways.The proposed framework employs wavelet transform to unravel the periodic characteristics of wind speed time series and integrates multi-source data(including ERA5 wind datasets,sand samples,ASTER GDEM,and multi-temporal remote sensing imagery)to enable a comprehensive aeolian environmental assessment.Concurrently,a suite of adaptive strategies is formulated to mitigate disaster risks along desert highways.Validated through a case study of the Tumushuk-Kunyu Desert Highway in Xinjiang,China,the framework exhibits high accuracy:predictions of annual aeolian sand transport activity show relative errors mostly below 7%against long-term reference sequences,and the calculated resultant drift direction exhibits a strong correlation with observed dune migration,yielding an R-squared value of 0.96.These findings confirm the framework’s reliability and provide a robust basis for designing adaptive,location-specific mitigation strategies,thereby enhancing the sustainability of desert highway infrastructure.展开更多
Based on the Gauss linear frequency modulated wavelet transform, a new characteristic index is presented, namely time frequency energy attenuation factor which can reflect the difference features of waveform in earthq...Based on the Gauss linear frequency modulated wavelet transform, a new characteristic index is presented, namely time frequency energy attenuation factor which can reflect the difference features of waveform in earthquake focus mechanism, wave traveling path and its attenuation characteristics in focal area or near field. In order to test its validity, we select the natural earthquakes and explosion or collapse events whose focus mechanisms vary obviously,and some natural earthquakes located at the same site or in a very small area. The study indicates that the time frequency energy attenuation factors of the natural earthquakes are obviously different with that of explosion or collapse events, and the change of the time frequency energy attenuation factors is relatively stable for the earthquakes under the normal seismicity background. Using the above mentioned method, it is expected to offer a useful criterion for strong earthquake prediction by continuous earthquake observation.展开更多
基金supported by the Innovative Human Resource Development for Local Intel-lectualization program through the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(No.IITP-2026-2020-0-01741)the research fund of Hanyang University(HY-2025-1110).
文摘Arrhythmias are a frequently occurring phenomenon in clinical practice,but how to accurately dis-tinguish subtle rhythm abnormalities remains an ongoing difficulty faced by the entire research community when conducting ECG-based studies.From a review of existing studies,two main factors appear to contribute to this problem:the uneven distribution of arrhythmia classes and the limited expressiveness of features learned by current models.To overcome these limitations,this study proposes a dual-path multimodal framework,termed DM-EHC(Dual-Path Multimodal ECG Heartbeat Classifier),for ECG-based heartbeat classification.The proposed framework links 1D ECG temporal features with 2D time–frequency features.By setting up the dual paths described above,the model can process more dimensions of feature information.The MIT-BIH arrhythmia database was selected as the baseline dataset for the experiments.Experimental results show that the proposed method outperforms single modalities and performs better for certain specific types of arrhythmias.The model achieved mean precision,recall,and F1 score of 95.14%,92.26%,and 93.65%,respectively.These results indicate that the framework is robust and has potential value in automated arrhythmia classification.
基金supported by the National Natural Science Foundation of China(Grant Nos.52379098,52274075)the Project of Xingliao Talents Program(XLYC2203008)the Science and Technology Program Project of Liaoning Province(2025JH2/101900011).
文摘Understanding how rock slopes respond to blasting loads is crucial for maintaining excavation safety and slope stability.Nevertheless,the spatiotemporal evolution,nonlinear dependence on blasting parameters,and predictive behavior of dominant frequency responses in slope vibrations remain insufficiently understood and quantified.This study combines time-frequency analysis with machine learning to explore how the dominant frequency(f_(d))evolves in slopes under blasting.Continuous Wavelet Transform(CWT)was employed to characterize the temporal-frequency evolution of vibration signals,revealing that the dominant frequency exhibits strong spatial dependence and nonlinear variability influenced by blasting parameters and rock mass structures.Three machine learning models,namely Back Propagation Neural Network(BP),Support Vector Machine(SVM),and Random Forest(RF),were developed to predict f_(d) based on 1,000 monitoring samples obtained from numerical and field simulations.Among them,the RF model achieved the highest prediction accuracy,with mean absolute percentage errors(MAPE)below 15%,demonstrating strong robustness and generalization capability.Our analysis shows that external excitation factors,especially the loading frequency(f_(d)),mainly control the frequency response,while internal controlling factors,such as spatial position,lithological variation,and mechanical heterogeneity,modulate localized frequency amplification and energy redistribution.The results reveal that f_(d) tends to decrease with elevation and distance from the blasting source,whereas structural planes and weathered zones induce high-frequency amplification due to scattering and modal coupling effects.This study offers a new framework combining time-frequency analysis and machine learning to measure the nonlinear interaction between blasting and rock mass response,offering new insights for dynamic stability evaluation and hazard mitigation in complex rock slope systems.
基金National Natural Science Foundation of China(Grant Nos.42388102,42030105,42192535)the Open Fund of State Key Laboratory of Precision Geodesy,Innovation Academy for Precision Measurement Science and Technology,Chinese Academy of Sciences(Grant No.SKLPG2025-1-5)。
文摘The state-of-the-art optical atomic clocks and the time-frequency signal transmission open a fresh field for gravity potential(geopotential)determination.Various methods,including optical fiber frequency transfer,satellite two-way,satellite common-view,satellite carrier phase,VLBI,tri-frequency combination,and dual-frequency combination,were developed to determine the geopotential differences using optical atomic clocks and then determine the geopotential at station B based on the geopotential at station A.This review elaborates the principles,methods,scientific objectives,applications,and relevant research trends of geopotential determination based on time-frequency signals.
基金supported in part by the National Natural Science Foundation of China(62372385,62272078,62002337)Chongqing Natural Science Foundation(CSTB2022NSCQ-MSX1486,CSTB2023NSCQ-LZX0069)。
文摘A dynamic graph(DG)is adopted to portray the evolving interplay between nodes in real-world scenarios prevalently.A high-order graph convolutional network(HGCN)is equipped with the ability to represent a DG by the spatial-temporal message passing mechanism built on tensor product.Concretely,an HGCN utilizes the discrete Fourier transform(DFT)to implement temporal message passing and then employs face-wise product to realize spatial message passing.However,DFT is only a special case of assorted time-frequency transforms,which considers the complex temporal patterns partially,thereby resulting in an inaccurate temporal message passing possibly.To address this issue,this study proposes six advanced time-frequency transform-incorporated HGCNs(TF-HGCNs)with discrete Fourier,discrete Hartley,discrete cosine,Haar wavelet,Walsh Hadamard,and slant transforms.In addition,a potent ensemble is built regarding the proposed six TF-HGCNs as the bases.Finally,the corresponding theoretical proof is presented.Empirical studies on six DG datasets demonstrate that owing to diverse time-frequency transforms,the proposed six TF-HGCNs significantly outperform state-of-the-art models in addressing the task of link weight estimation.Moreover,their ensemble outstrips each base's performance.
基金supported in part by the National Natural Science Foundation of China Grant 11962006the Natural Science Foundation of Jiangxi Province of China Grant 20232BAB204067.
文摘Quantitative detection of sleeve grouting compactness is a technical challenge in civil engineering testing.This study explores a novel quantitative detection method based on ultrasonic time-frequency dual-domain analysis.It establishes a mapping relationship between sleeve grouting compactness and characteristic parameters.First,this study made samples with gradient defects for two types of grouting sleeves,G18 and G20.These included four cases:2D,4D,6D defects(where D is the diameter of the grouting sleeve),and no-defect.Then,an ultrasonic input/output data acquisition system was established.Three-dimensional sound field distribution data were obtained through an orthogonal detection layout and pulse reflection principles.Finally,a novel quantification detection with a comprehensive defect index(DI)was established by comprehensively considering eight feature parameters,such as time-frequency domain Kurtosis factor(KU),Skewness factor(SK),Formfactor(FF),Crest factor(CF),Impulse factor(IF),Clearance factor(CLF),Wavelet packet energy entropy(WPEE),and Hilbert energy peak(HEP).Construct a DI index by quantifying the difference between defect signals and defect free signals in the time-frequency domain.Experimental results show that,under no-defect conditions,the values of feature parameters are significantly lower than those under defect conditions.Among these,the KU,FF,CF,WPEE and HEP exhibit strong correlations with grout sleeve compactness.The proposed DI index in both types of grout sleeves showed good universality with a linear fit goodness of 0.847–0.962.However,G20 the larger inner diameter and length of the sleeve result in a more complex medium effect during ultrasonic propagation,making its DI index more sensitive to defects than the G18 sleeve.Therefore,the presented method is effective for quantitative detection and analysis of the compactness of grouting sleeves.
文摘Fast wavelet multi-resolution analysis (wavelet MRA) provides a effective tool for analyzing and canceling disturbing components in original signal. Because of its exponential frequency axis, this method isn't suitable for extracting harmonic components. The modified exponential time-frequency distribution ( MED) overcomes the problems of Wigner distribution( WD) ,can suppress cross-terms and cancel noise further more. MED provides high resolution in both time and frequency domains, so it can make out weak period impulse components fmm signal with mighty harmonic components. According to the 'time' behavior, together with 'frequency' behavior in one figure,the essential structure of a signal is revealed clearly. According to the analysis of algorithm and fault diagnosis example, the joint of wavelet MRA and MED is a powerful tool for fault diagnosis.
文摘A critical problem facing data collection in structural health monitoring,for instance via sensor networks,is how to extract the main components and useful features for damage detection.A structural dynamic measurement is more often a complex time-varying process and therefore,is prone to dynamic changes in time-frequency contents.To extract the signal components and capture the useful features associated with damage from such nonstationary signals,a technique that combines the time and frequency analysis and shows the signal evolution in both time and frequency is required.Wavelet analyses have proven to be a viable and effective tool in this regard.Wavelet transform(WT)can analyze different signal components and then comparing the characteristics of each signal with a resolution matched to its scale.However,the challenge is the selection of a proper wavelet since various wavelets with varied properties that are to analyze the same data may result in different results.This article presents a study on how to carry out a comparative analysis based on analytic wavelet scalograms,using structural dynamic acceleration responses,to evaluate the effectiveness of various wavelets for damage detection in civil structures.The scalogram’s informative time-frequency regions are examined to analyze the variation of wavelet coefficients and show how the frequency content of a signal changes over time to detect transient events due to damage.Subsequently,damage-induced changes are tracked with time-frequency representations.Towards this aim,energy distribution and sharing information are investigated.The undamaged and damaged simulated comparative results of a structure reveal that the damaged structure were shifted from the undamaged structure.Also,the Bump wavelet shows the best results than the others.
文摘In the time-frequency analysis of seismic signals, the matching pursuit algorithm is an effective tool for non-stationary signals, and has high time-frequency resolution and a transient structure with local self-adaption. We expand the time-frequency dictionary library with Ricker, Morlet, and mixed phase seismic wavelets, to make the method more suitable for seismic signal time-frequency decomposition. In this paper, we demonstrated the algorithm theory using synthetic seismic data, and tested the method using synthetic data with 25% noise. We compared the matching pursuit results of the time-frequency dictionaries. The results indicated that the dictionary which matched the signal characteristics better would obtain better results, and can reflect the information of seismic data effectively.
基金Supported by National Natural Science Foundation of China (Grant Nos.52101343,52201351)。
文摘Vibration signals have the characteristics of multi-source strong shock coupling and strong noise interference owing to the complex structure of reciprocating machinery.Therefore,it is difficult to extract,analyze,and diagnose mechanical fault features.To accurately extract sensitive features from the strong noise interference and unsteady monitoring signals of reciprocating machinery,a study on the time-frequency feature extraction method of multi-source shock signals is conducted.Combining the characteristics of reciprocating mechanical vibration signals,a targeted optimization method considering the variational modal decomposition(VMD)mode number and second penalty factor is proposed,which completed the adaptive decomposition of coupled signals.Aiming at the bilateral asymmetric attenuation characteristics of reciprocating mechanical shock signals,a new bilateral adaptive Laplace wavelet(BALW)is established.A search strategy for wavelet local parameters of multi-shock signals is proposed using the harmony search(HS)method.A multi-source shock simulation signal is established,and actual data on the valve fault are obtained through diesel engine fault experiments.The fault recognition rate of the intake and exhaust valve clearance is above 90%and the extraction accuracy of the shock start position is improved by 10°.
基金This work has been supported by the following:the National Natural Science Foundation of China(No.91958206,91858215)the National Key Research and Development Program Pilot Project(No.2018YFC1405901,2017YFC0307401)+1 种基金the Fundamental Research Funds for the Central Universities(No.201964016)the Marine Geological Survey Program of China Geological Survey(No.DD20190819).
文摘An air gun generates acoustic signals for seismic exploration by releasing a high-pressure gas.A large error is always gradually introduced into the ideal-gas model when the pressure in the air-gun chamber exceeds 100 atm.In the van der Waals non-ideal-gas theory,the gas in the air gun can be regarded as an actual gas,and the error is less than 2%.The van der Waals model is established in combination with the quasi-static open thermodynamic system and bubble-motion equation by considering the bubble rise,bubble interaction,and throttling eff ect.The mismatch between the van der Waals and ideal-gas models is related to the pressure.Theoretically,under high-pressure conditions,the van der Waals air-gun model yields results that are closer to the measured results.Marine vertical cables are extended to the seafl oor using steel cables that connect the cement blocks,but the corresponding hydrophones are suspended in the seawater.Thus,noise associated with ships,ocean surges,and coupling problems is avoided,and the signal-to-noise ratio and resolution of marine seismic data are improved.This acquisition method satisfies the conditions of recording air-gun far-fi eld wavelets.According to an actual vertical-cable observation system,the van der Waals air-gun model is used to model the wavelet of different azimuth and take-off angles.The characteristics of the experimental and simulated data demonstrate good agreement,which indicates that the van der Waals method is accurate and reliable.The accuracy of the model is directly related to the resolution,thus aff ecting the resolution ability of the stratum.
基金jointly sponsored the Special Fund for Earthquake Scientific Research of China Earthquake Administration(2015419015)the National Natural Sciences Foundation of China(41474071)
文摘Through analyzing the near-field hydrophone records of the airgun experiment in the Jiemian reservoir,Fujian,we study the time-frequency characteristic of airgun source wavelet and the influence of gun depth and firing pressure,and explain the process of bubble oscillation based on the Johnson( 1994) bubble model. The data analysis shows that:( 1) Airgun wavelet is composed of primary pulse and bubble pulse. The primary pulse,which is of large amplitude,short duration and wide frequency band,is usually used in shallow exploration. The bubble pulse,which is concentrated in the low-frequency range,is usually used in deep exploration with deep vertical penetration and far horizontal propagation.( 2) The variation of primary pulse amplitude with gun depth is very small,bubble pulse amplitude and the dominant frequency increase,and peak-bubble ratio and bubble period decrease. When the gun depth is 10 m,primary pulse amplitude and peakbubble ratio are maximum,which is suitable for shallow exploration; when gun depth is25 m,bubble pulse amplitude is large, and peak-bubble ratio is minimum, which is suitable for deep exploration.( 3) The primary pulse amplitude,bubble pulse amplitude,peak-bubble ratio,and bubble period increase and the dominant frequency decreases with increased firing pressure.
基金supported by Natural Science Foundation of China(No.62371231)Natural Science Foundation on Frontier Leading Technology Basic Research Project of Jiangsu under Grant BK20222001Jiangsu Provincial Key Research and Development Program(No.BE2023027).
文摘With the increasingly complex and changeable electromagnetic environment,wireless communication systems are facing jamming and abnormal signal injection,which significantly affects the normal operation of a communication system.In particular,the abnormal signals may emulate the normal signals,which makes it very challenging for abnormal signal recognition.In this paper,we propose a new abnormal signal recognition scheme,which combines time-frequency analysis with deep learning to effectively identify synthetic abnormal communication signals.Firstly,we emulate synthetic abnormal communication signals including seven jamming patterns.Then,we model an abnormal communication signals recognition system based on the communication protocol between the transmitter and the receiver.To improve the performance,we convert the original signal into the time-frequency spectrogram to develop an image classification algorithm.Simulation results demonstrate that the proposed method can effectively recognize the abnormal signals under various parameter configurations,even under low signal-to-noise ratio(SNR)and low jamming-to-signal ratio(JSR)conditions.
文摘The local wave method is a very good time-frequency method for nonstationaryvibration signal analysis. But the interfering noise has a big influence on the accuracy oftime-frequency analysis. The wavelet packet de-noising method can eliminate the interference ofnoise and improve the signal-noise-ratio. This paper uses the local wave method to decompose thede-noising signal and perform a time-frequency analysis. We can get better characteristics. Finally,an example of wavelet packet de-noising and a local wave time-frequency spectrum application ofdiesel engine surface vibration signal is put forward.
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funded by National Social Science Fund of China,grant number 23BYY197.
文摘Image captioning,a pivotal research area at the intersection of image understanding,artificial intelligence,and linguistics,aims to generate natural language descriptions for images.This paper proposes an efficient image captioning model named Mob-IMWTC,which integrates improved wavelet convolution(IMWTC)with an enhanced MobileNet V3 architecture.The enhanced MobileNet V3 integrates a transformer encoder as its encoding module and a transformer decoder as its decoding module.This innovative neural network significantly reduces the memory space required and model training time,while maintaining a high level of accuracy in generating image descriptions.IMWTC facilitates large receptive fields without significantly increasing the number of parameters or computational overhead.The improvedMobileNet V3 model has its classifier removed,and simultaneously,it employs IMWTC layers to replace the original convolutional layers.This makes Mob-IMWTC exceptionally well-suited for deployment on lowresource devices.Experimental results,based on objective evaluation metrics such as BLEU,ROUGE,CIDEr,METEOR,and SPICE,demonstrate that Mob-IMWTC outperforms state-of-the-art models,including three CNN architectures(CNN-LSTM,CNN-Att-LSTM,CNN-Tran),two mainstream methods(LCM-Captioner,ClipCap),and our previous work(Mob-Tran).Subjective evaluations further validate the model’s superiority in terms of grammaticality,adequacy,logic,readability,and humanness.Mob-IMWTC offers a lightweight yet effective solution for image captioning,making it suitable for deployment on resource-constrained devices.
基金supported in part by the National Natural Science Foundation of China[Grant number 62471075]the Major Science and Technology Project Grant of the Chongqing Municipal Education Commission[Grant number KJZD-M202301901].
文摘Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at https://github.com/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).
基金supported by the National Natural Science Foundation of China(42430303)Strategy Priority Research Program(Category B)of the Chinese Academy of Sciences(XDB0710000)+2 种基金National Natural Science Foundation of China(42288201)the National Key R&D Program of China(2023YFF0803203)the IGGCAS start-up funding(Grant No.E251510101).
文摘In wave-equation migration and demigration,the cross-correlation imaging/forwarding step implicitly injects an additional copy of the source wavelet,so that the amplitude spectrum of the wavelet is applied redundantly(effectively imposing a wavelet-spectrum weighting,often akin to an amplitude-squared bias).This redundancy degrades structural fidelity and amplitude balance yet is frequently overlooked.We(i)formalize the mechanism by which cross-correlation duplicates the source-wavelet amplitude effect in both migration and demigration,and(ii)introduce a source-equalized operator that removes the redundancy by deconvolving(or dividing by)the wavelet amplitude spectrum in the imaging condition and its demigration counterpart,while leaving phase/kinematics intact.Using a band-limited Ricker wavelet on a two-layer model and on Marmousi,we show that,if unmanaged,the redundant wavelet spectrum broadens main lobes,introduces ringing,and suppresses vertical resolution in migrated images,and inflates spectrum mismatches between demigrated and observed data even when peak times agree.With our correction,images recover observed-data-consistent bandwidth and sharpened interfaces,and demigrated data also exhibit improved spectrum conformity and reduced amplitude misfit.The results clarify when source amplitudes matter,why cross-correlation makes them redundantly matter,and how a lightweight spectral correction restores physically meaningful amplitude behavior in wave-equation migration/demigration.
基金jointly funded by the Joint Funds of the National Natural Science Foundation of China(Grant No.U2568210)the Interdisciplinary Research Program of Shihezi University(Grant No.JCYJ202317)the National Natural Science Foundation of China(Grant No.12362035)。
文摘Scientific analysis of aeolian sand environments is fundamental for sustainable disaster mitigation along desert highways.However,significant regional variability in wind energy conditions complicates accurate characterization of wind regimes and introduces uncertainty in determining optimal monitoring timescales.Moreover,prevailing sand control measures often rely on standardized designs rather than site-specific adaptive strategies.To address these issues,this study proposes an integrated framework for aeolian environment analysis and develops targeted disaster mitigation strategies tailored for desert highways.The proposed framework employs wavelet transform to unravel the periodic characteristics of wind speed time series and integrates multi-source data(including ERA5 wind datasets,sand samples,ASTER GDEM,and multi-temporal remote sensing imagery)to enable a comprehensive aeolian environmental assessment.Concurrently,a suite of adaptive strategies is formulated to mitigate disaster risks along desert highways.Validated through a case study of the Tumushuk-Kunyu Desert Highway in Xinjiang,China,the framework exhibits high accuracy:predictions of annual aeolian sand transport activity show relative errors mostly below 7%against long-term reference sequences,and the calculated resultant drift direction exhibits a strong correlation with observed dune migration,yielding an R-squared value of 0.96.These findings confirm the framework’s reliability and provide a robust basis for designing adaptive,location-specific mitigation strategies,thereby enhancing the sustainability of desert highway infrastructure.
文摘Based on the Gauss linear frequency modulated wavelet transform, a new characteristic index is presented, namely time frequency energy attenuation factor which can reflect the difference features of waveform in earthquake focus mechanism, wave traveling path and its attenuation characteristics in focal area or near field. In order to test its validity, we select the natural earthquakes and explosion or collapse events whose focus mechanisms vary obviously,and some natural earthquakes located at the same site or in a very small area. The study indicates that the time frequency energy attenuation factors of the natural earthquakes are obviously different with that of explosion or collapse events, and the change of the time frequency energy attenuation factors is relatively stable for the earthquakes under the normal seismicity background. Using the above mentioned method, it is expected to offer a useful criterion for strong earthquake prediction by continuous earthquake observation.
