Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approach...Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.展开更多
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
This paper introduces a novel method for medical image retrieval and classification by integrating a multi-scale encoding mechanism with Vision Transformer(ViT)architectures and a dynamic multi-loss function.The multi...This paper introduces a novel method for medical image retrieval and classification by integrating a multi-scale encoding mechanism with Vision Transformer(ViT)architectures and a dynamic multi-loss function.The multi-scale encoding significantly enhances the model’s ability to capture both fine-grained and global features,while the dynamic loss function adapts during training to optimize classification accuracy and retrieval performance.Our approach was evaluated on the ISIC-2018 and ChestX-ray14 datasets,yielding notable improvements.Specifically,on the ISIC-2018 dataset,our method achieves an F1-Score improvement of+4.84% compared to the standard ViT,with a precision increase of+5.46% for melanoma(MEL).On the ChestX-ray14 dataset,the method delivers an F1-Score improvement of 5.3%over the conventional ViT,with precision gains of+5.0% for pneumonia(PNEU)and+5.4%for fibrosis(FIB).Experimental results demonstrate that our approach outperforms traditional CNN-based models and existing ViT variants,particularly in retrieving relevant medical cases and enhancing diagnostic accuracy.These findings highlight the potential of the proposedmethod for large-scalemedical image analysis,offering improved tools for clinical decision-making through superior classification and case comparison.展开更多
The capacity to diagnose faults in rolling bearings is of significant practical importance to ensure the normal operation of the equipment.Frequency-domain features can effectively enhance the identification of fault ...The capacity to diagnose faults in rolling bearings is of significant practical importance to ensure the normal operation of the equipment.Frequency-domain features can effectively enhance the identification of fault modes.However,existing methods often suffer from insufficient frequency-domain representation in practical applications,which greatly affects diagnostic performance.Therefore,this paper proposes a rolling bearing fault diagnosismethod based on aMulti-Scale FusionNetwork(MSFN)using the Time-Division Fourier Transform(TDFT).The method constructs multi-scale channels to extract time-domain and frequency-domain features of the signal in parallel.A multi-level,multi-scale filter-based approach is designed to extract frequency-domain features in a segmented manner.A cross-attention mechanism is introduced to facilitate the fusion of the extracted time-frequency domain features.The performance of the proposed method is validated using the CWRU and Ottawa datasets.The results show that the average accuracy of MSFN under complex noisy signals is 97.75%and 94.41%.The average accuracy under variable load conditions is 98.68%.This demonstrates its significant application potential compared to existing methods.展开更多
Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete v...Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete vessel segmentation and poor continuity.In this study,we propose CT-MFENet to address the aforementioned issues.First,the use of context transformer(CT)allows for the integration of contextual feature information,which helps establish the connection between pixels and solve the problem of incomplete vessel continuity.Second,multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales.In the decoding stage,we introduce a local-global fusion module.It enhances the localization of vascular information and reduces the semantic gap between high-and low-level features.To address the class imbalance in retinal images,we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures.We conducted experiments on the publicly available DRIVE,CHASEDB1,STARE,and IOSTAR datasets.The experimental results show that our CT-MFENet performs better than most existing methods,including the baseline U-Net.展开更多
Noise has traditionally been suppressed or eliminated in seismic data sets by the use of Fourier filters and, to a lesser degree, nonlinear statistical filters. Although these methods are quite useful under specific c...Noise has traditionally been suppressed or eliminated in seismic data sets by the use of Fourier filters and, to a lesser degree, nonlinear statistical filters. Although these methods are quite useful under specific conditions, they may produce undesirable effects for the low signal to noise ratio data. In this paper, a new method, multi-scale ridgelet transform, is used in the light of the theory of ridgelet transform. We employ wavelet transform to do sub-band decomposition for the signals and then use non-linear thresholding in ridgelet domain for every block. In other words, it is based on the idea of partition, at sufficiently fine scale, a curving singularity looks straight, and so ridgelet transform can work well in such cases. Applications on both synthetic data and actual seismic data from Sichuan basin, South China, show that the new method eliminates the noise portion of the signal more efficiently and retains a greater amount of geologic data than other methods, the quality and consecutiveness of seismic event are improved obviously as well as the quality of section is improved.展开更多
A novel model named Multi-scale Gaussian Processes (MGP) is proposed. Motivated by the ideas of multi-scale representations in the wavelet theory, in the new model, a Gaussian process is represented at a scale by a li...A novel model named Multi-scale Gaussian Processes (MGP) is proposed. Motivated by the ideas of multi-scale representations in the wavelet theory, in the new model, a Gaussian process is represented at a scale by a linear basis that is composed of a scale function and its different translations. Finally the distribution of the targets of the given samples can be obtained at different scales. Compared with the standard Gaussian Processes (GP) model, the MGP model can control its complexity conveniently just by adjusting the scale pa-rameter. So it can trade-off the generalization ability and the empirical risk rapidly. Experiments verify the fea-sibility of the MGP model, and exhibit that its performance is superior to the GP model if appropriate scales are chosen.展开更多
Infrared-visible image fusion plays an important role in multi-source data fusion,which has the advantage of integrating useful information from multi-source sensors.However,there are still challenges in target enhanc...Infrared-visible image fusion plays an important role in multi-source data fusion,which has the advantage of integrating useful information from multi-source sensors.However,there are still challenges in target enhancement and visual improvement.To deal with these problems,a sub-regional infrared-visible image fusion method(SRF)is proposed.First,morphology and threshold segmentation is applied to extract targets interested in infrared images.Second,the infrared back-ground is reconstructed based on extracted targets and the visible image.Finally,target and back-ground regions are fused using a multi-scale transform.Experimental results are obtained using public data for comparison and evaluation,which demonstrate that the proposed SRF has poten-tial benefits over other methods.展开更多
BACKGROUND： Recent studies have focused on various methods of wavelet transformation for electroencephalogram （EEG） signals. However, there are very few studies reporting characteristics of multi-scale phase waves ...BACKGROUND： Recent studies have focused on various methods of wavelet transformation for electroencephalogram （EEG） signals. However, there are very few studies reporting characteristics of multi-scale phase waves during epileptic discharge.OBJECTIVE： To extract multi-scale phase average waveforms from childhood absence epilepsy EEG signals between time and frequency domains using wavelet transformation, and to compare EEG signals of absence seizure with pre-epileptic seizure and normal children, and to quantify multi-scale phase average waveforms from childhood absence epilepsy EEG signals. DESIGN, TIME AND SETTING： The case-comparative experiment was performed at the Department of Neuroelectrophysiology, Tianjin Medical University from August 2002 to May 2005. PARTICIPANTS： A total of 15 patients with childhood absence epilepsy from the General Hospital of Tianjin Medical University were enrolled in the study. The patients were not administered anti-epileptic drugs or sedatives prior to EEG testing. In addition, 12 healthy, age- and gender-matched children were also enrolled.METHODS： EEG signals were tested on 15 patients with childhood absence epilepsy and 12 normal children. Epileptic discharge signals during clinical and subclinical seizures were collected 10 and 20 times, respectively. The collected EEG signals were treated with wavelet transformation to extract multi-scale characteristics during absence epilepsy seizure using a conditional sampling method. Multi-scale phase average waveforms were collected using a conditional phase averaging technique. Amplitude of phase average waveform from EEG signals of epilepsy seizure, subclinical epileptic discharge, and EEG signals of normal children were compared and statistically analyzed in the first half-cycle.MAIN OUTCOME MEASURES： Multi-scale wavelet coefficient and the evolution of EEG signals were observed during childhood absence epilepsy seizures using wavelet transformation. Multi-scale phase average waveforms from EEG signals were observed using a conditional sampling method and phase averaging technique.RESULTS： Multi-scale characteristics of EEG signals demonstrated that 12-scale （3 Hz） rhythmical activity was significantly enhanced during childhood absence epilepsy seizure and co-existed with background structure （〈1 Hz, low frequency discharge）. The phase average wave exhibited opposed phase abnormal rhythm at 3 Hz. Prior to childhood absence epilepsy seizure, EEG detected opposed abnormal a rhythm and 3 Hz composition, which were not detected with traditional EEG. Compared to EEG signals from normal children, epileptic discharges from clinical and subclinical childhood absence epilepsy seizures were positive and amplitude was significantly greater （P〈0.05）.CONCLUSION： Wavelet transformation was used to analyze EEG signals from childhood absence epilepsy to obtain multi-scale quantitative characteristics and phase average waveforms. Multi-scale wavelet coefficients of EEG signals correlated with childhood absence epilepsy seizure, and multi-scale waveforms prior to epilepsy seizure were similar to characteristics during the onset period. Compared to normal children, EEG signals during epilepsy seizure exhibited an opposed phase model.展开更多
The high-frequency components in the traditional multi-scale transform method are approximately sparse, which can represent different information of the details. But in the low-frequency component, the coefficients ar...The high-frequency components in the traditional multi-scale transform method are approximately sparse, which can represent different information of the details. But in the low-frequency component, the coefficients around the zero value are very few, so we cannot sparsely represent low-frequency image information. The low-frequency component contains the main energy of the image and depicts the profile of the image. Direct fusion of the low-frequency component will not be conducive to obtain highly accurate fusion result. Therefore, this paper presents an infrared and visible image fusion method combining the multi-scale and top-hat transforms. On one hand, the new top-hat-transform can effectively extract the salient features of the low-frequency component. On the other hand, the multi-scale transform can extract highfrequency detailed information in multiple scales and from diverse directions. The combination of the two methods is conducive to the acquisition of more characteristics and more accurate fusion results. Among them, for the low-frequency component, a new type of top-hat transform is used to extract low-frequency features, and then different fusion rules are applied to fuse the low-frequency features and low-frequency background; for high-frequency components, the product of characteristics method is used to integrate the detailed information in high-frequency. Experimental results show that the proposed algorithm can obtain more detailed information and clearer infrared target fusion results than the traditional multiscale transform methods. Compared with the state-of-the-art fusion methods based on sparse representation, the proposed algorithm is simple and efficacious, and the time consumption is significantly reduced.展开更多
Segmenting skin lesions is critical for early skin cancer detection.Existing CNN and Transformer-based methods face challenges such as high computational complexity and limited adaptability to variations in lesion siz...Segmenting skin lesions is critical for early skin cancer detection.Existing CNN and Transformer-based methods face challenges such as high computational complexity and limited adaptability to variations in lesion sizes.To overcome these limitations,we introduce MSAMamba-UNet,a lightweight model that integrates two novel architectures:Multi-Scale Mamba(MSMamba)and Adaptive Dynamic Gating Block(ADGB).MSMamba utilizes multi-scale decomposition and a parallel hierarchical structure to enhance the delineation of irregular lesion boundaries and sensitivity to small targets.ADGB dynamically selects convolutional kernels with varying receptive fields based on input features,improving the model’s capacity to accommodate diverse lesion textures and scales.Additionally,we introduce a Mix Attention Fusion Block(MAF)to enhance shallow feature representation by integrating parallel channel and pixel attention mechanisms.Extensive evaluation of MSAMamba-UNet on the ISIC 2016,ISIC 2017,and ISIC 2018 datasets demonstrates competitive segmentation accuracy with only 0.056 M parameters and 0.069 GFLOPs.Our experiments revealed that MSAMamba-UNet achieved IoU scores of 85.53%,85.47%,and 82.22%,as well as DSC scores of 92.20%,92.17%,and 90.24%,respectively.These results underscore the lightweight design and effectiveness of MSAMamba-UNet.展开更多
When using a miniature single sensor boundary layer probe, the time sequences of the stream-wise velocity in the turbulent boundary layer （TBL） are measured by using a hot wire anemometer. Beneath the fully develope...When using a miniature single sensor boundary layer probe, the time sequences of the stream-wise velocity in the turbulent boundary layer （TBL） are measured by using a hot wire anemometer. Beneath the fully developed TBL, the wall pressure fluctuations are attained by a microphone mechanism with high spatial resolution. Analysis on the statistic and spectrum properties of velocity and wall pressure reveals the relationship between the wall pressure fluctuation and the energy-containing structure in the buffer layer of the TBL. Wavelet transform shows the multi-scale natures of coherent structures contained in both signals of velocity and pressure. The most intermittent wall pressure scale is associated with the coherent structure in the buffer layer. Meanwhile the most energetic scale of velocity fluctuation at y＋ = 14 provides a specific frequency f9 ≈ 147 Hz for wall actuating control with Ret = 996.展开更多
Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor c...Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor control variables are necessary.Common data assimilation systems theoretically require that the probability density functions(PDFs)of analysis,background,and observation errors should satisfy the Gaussian unbiased assumptions.In this study,a Gaussian transform method is proposed to transform hydrometeors to more Gaussian variables,which is modified from the Softmax function and renamed as Quasi-Softmax transform.The Quasi-Softmax transform method then is compared to the original hydrometeor mixing ratios and their logarithmic transform and Softmax transform.The spatial distribution,the non-Gaussian nature of the background errors,and the characteristics of the background errors of hydrometeors in each method are studied.Compared to the logarithmic and Softmax transform,the Quasi-Softmax method keeps the vertical distribution of the original hydrometeor mixing ratios to the greatest extent.The results of the D′Agostino test show that the hydrometeors transformed by the Quasi-Softmax method are more Gaussian when compared to the other methods.The Gaussian transform has been added to the control variable transform to estimate the background error covariances.Results show that the characteristics of the hydrometeor background errors are reasonable for the Quasi-Softmax method.The transformed hydrometeors using the Quasi-Softmax transform meet the Gaussian unbiased assumptions of the data assimilation system,and are promising control variables for data assimilation systems.展开更多
Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.Th...Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.This paper presents a convolutional structure with multi-scale fusion to optimize the step of clothing feature extraction and a self-attention module to capture long-range association information.The structure enables the self-attention mechanism to directly participate in the process of information exchange through the down-scaling projection operation of the multi-scale framework.In addition,the improved self-attention module introduces the extraction of 2-dimensional relative position information to make up for its lack of ability to extract spatial position features from clothing images.The experimental results based on the colorful fashion parsing dataset(CFPD)show that the proposed network structure achieves 53.68%mean intersection over union(mIoU)and has better performance on the clothing parsing task.展开更多
Feature extraction of signals plays an important role in classification problems because of data dimension reduction property and potential improvement of a classification accuracy rate. Principal component analysis (...Feature extraction of signals plays an important role in classification problems because of data dimension reduction property and potential improvement of a classification accuracy rate. Principal component analysis (PCA), wavelets transform or Fourier transform methods are often used for feature extraction. In this paper, we propose a multi-scale PCA, which combines discrete wavelet transform, and PCA for feature extraction of signals in both the spatial and temporal domains. Our study shows that the multi-scale PCA combined with the proposed new classification methods leads to high classification accuracy for the considered signals.展开更多
Laser transformation hardening(LTH) is one of the laser surface modification processes.The surface hardening of rod-shaped carbon steel(SM45C) was performed by lathe-based laser composite processor with Gaussian-beam ...Laser transformation hardening(LTH) is one of the laser surface modification processes.The surface hardening of rod-shaped carbon steel(SM45C) was performed by lathe-based laser composite processor with Gaussian-beam optical head.The LTH characteristics by dominant processes,longitudinal and depth directional hardness distributions and behaviors of phase transformation in hardened zones were examined.Especially,two concepts of circumferential speed and theoretical overlap rate were applied.When laser power increased or circumferential speed decreased,the surface hardening depth gradually increases due to the increased heat input.Moreover,the longitudinal hardness distribution particularly shows periodicity of repetitive increase and decrease,which results from tempering effect by overlap.Finally,the feasibility of laser transformation hardening is verified by using the beam with Gaussian intensity distribution.展开更多
To deal with the low location accuracy issue of existing underwater navigation technologies in autonomous underwater vehicles(AUVs),a distributed fusion algorithm which combines the model's analysis method with a ...To deal with the low location accuracy issue of existing underwater navigation technologies in autonomous underwater vehicles(AUVs),a distributed fusion algorithm which combines the model's analysis method with a multi-scale transformation method is proposed for integrated navigation system based on AUV.First,integrated navigation system theory and system error sources are introduced in details.Secondly,a navigation system's observation equation on the original scale is decomposed into different scales by the discrete wavelet transform method,and noise reduction is performed by setting the wavelet de-noising threshold.At last,the dynamic equation and observation equations are fused on different scales by the wavelet transformation and Kalman filter.The results show that the proposed algorithm has smaller navigation error and higher navigation accuracy.展开更多
Computer-aided diagnosis of pneumonia based on deep learning is a research hotspot.However,there are some problems that the features of different sizes and different directions are not sufficient when extracting the f...Computer-aided diagnosis of pneumonia based on deep learning is a research hotspot.However,there are some problems that the features of different sizes and different directions are not sufficient when extracting the features in lung X-ray images.A pneumonia classification model based on multi-scale directional feature enhancement MSD-Net is proposed in this paper.The main innovations are as follows:Firstly,the Multi-scale Residual Feature Extraction Module(MRFEM)is designed to effectively extract multi-scale features.The MRFEM uses dilated convolutions with different expansion rates to increase the receptive field and extract multi-scale features effectively.Secondly,the Multi-scale Directional Feature Perception Module(MDFPM)is designed,which uses a three-branch structure of different sizes convolution to transmit direction feature layer by layer,and focuses on the target region to enhance the feature information.Thirdly,the Axial Compression Former Module(ACFM)is designed to perform global calculations to enhance the perception ability of global features in different directions.To verify the effectiveness of the MSD-Net,comparative experiments and ablation experiments are carried out.In the COVID-19 RADIOGRAPHY DATABASE,the Accuracy,Recall,Precision,F1 Score,and Specificity of MSD-Net are 97.76%,95.57%,95.52%,95.52%,and 98.51%,respectively.In the chest X-ray dataset,the Accuracy,Recall,Precision,F1 Score and Specificity of MSD-Net are 97.78%,95.22%,96.49%,95.58%,and 98.11%,respectively.This model improves the accuracy of lung image recognition effectively and provides an important clinical reference to pneumonia Computer-Aided Diagnosis.展开更多
In the traditional incremental analysis update(IAU)process,all analysis increments are treated as constant forcing in a model’s prognostic equations over a certain time window.This approach effectively reduces high-f...In the traditional incremental analysis update(IAU)process,all analysis increments are treated as constant forcing in a model’s prognostic equations over a certain time window.This approach effectively reduces high-frequency oscillations introduced by data assimilation.However,as different scales of increments have unique evolutionary speeds and life histories in a numerical model,the traditional IAU scheme cannot fully meet the requirements of short-term forecasting for the damping of high-frequency noise and may even cause systematic drifts.Therefore,a multi-scale IAU scheme is proposed in this paper.Analysis increments were divided into different scale parts using a spatial filtering technique.For each scale increment,the optimal relaxation time in the IAU scheme was determined by the skill of the forecasting results.Finally,different scales of analysis increments were added to the model integration during their optimal relaxation time.The multi-scale IAU scheme can effectively reduce the noise and further improve the balance between large-scale and small-scale increments in the model initialization stage.To evaluate its performance,several numerical experiments were conducted to simulate the path and intensity of Typhoon Mangkhut(2018)and showed that:(1)the multi-scale IAU scheme had an obvious effect on noise control at the initial stage of data assimilation;(2)the optimal relaxation time for large-scale and small-scale increments was estimated as 6 h and 3 h,respectively;(3)the forecast performance of the multi-scale IAU scheme in the prediction of Typhoon Mangkhut(2018)was better than that of the traditional IAU scheme.The results demonstrate the superiority of the multi-scale IAU scheme.展开更多
基金funded by the National Natural Science Foundation of China,grant numbers 52374156 and 62476005。
文摘Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.
基金supported 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 the Deanship of Research and Graduate Studies at King Khalid University through small group research under grant number RGP1/278/45.
文摘This paper introduces a novel method for medical image retrieval and classification by integrating a multi-scale encoding mechanism with Vision Transformer(ViT)architectures and a dynamic multi-loss function.The multi-scale encoding significantly enhances the model’s ability to capture both fine-grained and global features,while the dynamic loss function adapts during training to optimize classification accuracy and retrieval performance.Our approach was evaluated on the ISIC-2018 and ChestX-ray14 datasets,yielding notable improvements.Specifically,on the ISIC-2018 dataset,our method achieves an F1-Score improvement of+4.84% compared to the standard ViT,with a precision increase of+5.46% for melanoma(MEL).On the ChestX-ray14 dataset,the method delivers an F1-Score improvement of 5.3%over the conventional ViT,with precision gains of+5.0% for pneumonia(PNEU)and+5.4%for fibrosis(FIB).Experimental results demonstrate that our approach outperforms traditional CNN-based models and existing ViT variants,particularly in retrieving relevant medical cases and enhancing diagnostic accuracy.These findings highlight the potential of the proposedmethod for large-scalemedical image analysis,offering improved tools for clinical decision-making through superior classification and case comparison.
基金fully supported by the Frontier Exploration Projects of Longmen Laboratory(No.LMQYTSKT034)Key Research and Development and Promotion of Special(Science and Technology)Project of Henan Province,China(No.252102210158)。
文摘The capacity to diagnose faults in rolling bearings is of significant practical importance to ensure the normal operation of the equipment.Frequency-domain features can effectively enhance the identification of fault modes.However,existing methods often suffer from insufficient frequency-domain representation in practical applications,which greatly affects diagnostic performance.Therefore,this paper proposes a rolling bearing fault diagnosismethod based on aMulti-Scale FusionNetwork(MSFN)using the Time-Division Fourier Transform(TDFT).The method constructs multi-scale channels to extract time-domain and frequency-domain features of the signal in parallel.A multi-level,multi-scale filter-based approach is designed to extract frequency-domain features in a segmented manner.A cross-attention mechanism is introduced to facilitate the fusion of the extracted time-frequency domain features.The performance of the proposed method is validated using the CWRU and Ottawa datasets.The results show that the average accuracy of MSFN under complex noisy signals is 97.75%and 94.41%.The average accuracy under variable load conditions is 98.68%.This demonstrates its significant application potential compared to existing methods.
基金the National Natural Science Foundation of China(No.62266025)。
文摘Segmentation of the retinal vessels in the fundus is crucial for diagnosing ocular diseases.Retinal vessel images often suffer from category imbalance and large scale variations.This ultimately results in incomplete vessel segmentation and poor continuity.In this study,we propose CT-MFENet to address the aforementioned issues.First,the use of context transformer(CT)allows for the integration of contextual feature information,which helps establish the connection between pixels and solve the problem of incomplete vessel continuity.Second,multi-scale dense residual networks are used instead of traditional CNN to address the issue of inadequate local feature extraction when the model encounters vessels at multiple scales.In the decoding stage,we introduce a local-global fusion module.It enhances the localization of vascular information and reduces the semantic gap between high-and low-level features.To address the class imbalance in retinal images,we propose a hybrid loss function that enhances the segmentation ability of the model for topological structures.We conducted experiments on the publicly available DRIVE,CHASEDB1,STARE,and IOSTAR datasets.The experimental results show that our CT-MFENet performs better than most existing methods,including the baseline U-Net.
基金supported by China Petrochemical key project during the 11th Five-year Plan as well as the Doctorate Fund of Ministry of Education of China (No.20050491504)
文摘Noise has traditionally been suppressed or eliminated in seismic data sets by the use of Fourier filters and, to a lesser degree, nonlinear statistical filters. Although these methods are quite useful under specific conditions, they may produce undesirable effects for the low signal to noise ratio data. In this paper, a new method, multi-scale ridgelet transform, is used in the light of the theory of ridgelet transform. We employ wavelet transform to do sub-band decomposition for the signals and then use non-linear thresholding in ridgelet domain for every block. In other words, it is based on the idea of partition, at sufficiently fine scale, a curving singularity looks straight, and so ridgelet transform can work well in such cases. Applications on both synthetic data and actual seismic data from Sichuan basin, South China, show that the new method eliminates the noise portion of the signal more efficiently and retains a greater amount of geologic data than other methods, the quality and consecutiveness of seismic event are improved obviously as well as the quality of section is improved.
文摘A novel model named Multi-scale Gaussian Processes (MGP) is proposed. Motivated by the ideas of multi-scale representations in the wavelet theory, in the new model, a Gaussian process is represented at a scale by a linear basis that is composed of a scale function and its different translations. Finally the distribution of the targets of the given samples can be obtained at different scales. Compared with the standard Gaussian Processes (GP) model, the MGP model can control its complexity conveniently just by adjusting the scale pa-rameter. So it can trade-off the generalization ability and the empirical risk rapidly. Experiments verify the fea-sibility of the MGP model, and exhibit that its performance is superior to the GP model if appropriate scales are chosen.
基金supported by the China Postdoctoral Science Foundation Funded Project(No.2021M690385)the National Natural Science Foundation of China(No.62101045).
文摘Infrared-visible image fusion plays an important role in multi-source data fusion,which has the advantage of integrating useful information from multi-source sensors.However,there are still challenges in target enhancement and visual improvement.To deal with these problems,a sub-regional infrared-visible image fusion method(SRF)is proposed.First,morphology and threshold segmentation is applied to extract targets interested in infrared images.Second,the infrared back-ground is reconstructed based on extracted targets and the visible image.Finally,target and back-ground regions are fused using a multi-scale transform.Experimental results are obtained using public data for comparison and evaluation,which demonstrate that the proposed SRF has poten-tial benefits over other methods.
基金the National Natural Science Foundation of China,No. 60703045
文摘BACKGROUND： Recent studies have focused on various methods of wavelet transformation for electroencephalogram （EEG） signals. However, there are very few studies reporting characteristics of multi-scale phase waves during epileptic discharge.OBJECTIVE： To extract multi-scale phase average waveforms from childhood absence epilepsy EEG signals between time and frequency domains using wavelet transformation, and to compare EEG signals of absence seizure with pre-epileptic seizure and normal children, and to quantify multi-scale phase average waveforms from childhood absence epilepsy EEG signals. DESIGN, TIME AND SETTING： The case-comparative experiment was performed at the Department of Neuroelectrophysiology, Tianjin Medical University from August 2002 to May 2005. PARTICIPANTS： A total of 15 patients with childhood absence epilepsy from the General Hospital of Tianjin Medical University were enrolled in the study. The patients were not administered anti-epileptic drugs or sedatives prior to EEG testing. In addition, 12 healthy, age- and gender-matched children were also enrolled.METHODS： EEG signals were tested on 15 patients with childhood absence epilepsy and 12 normal children. Epileptic discharge signals during clinical and subclinical seizures were collected 10 and 20 times, respectively. The collected EEG signals were treated with wavelet transformation to extract multi-scale characteristics during absence epilepsy seizure using a conditional sampling method. Multi-scale phase average waveforms were collected using a conditional phase averaging technique. Amplitude of phase average waveform from EEG signals of epilepsy seizure, subclinical epileptic discharge, and EEG signals of normal children were compared and statistically analyzed in the first half-cycle.MAIN OUTCOME MEASURES： Multi-scale wavelet coefficient and the evolution of EEG signals were observed during childhood absence epilepsy seizures using wavelet transformation. Multi-scale phase average waveforms from EEG signals were observed using a conditional sampling method and phase averaging technique.RESULTS： Multi-scale characteristics of EEG signals demonstrated that 12-scale （3 Hz） rhythmical activity was significantly enhanced during childhood absence epilepsy seizure and co-existed with background structure （〈1 Hz, low frequency discharge）. The phase average wave exhibited opposed phase abnormal rhythm at 3 Hz. Prior to childhood absence epilepsy seizure, EEG detected opposed abnormal a rhythm and 3 Hz composition, which were not detected with traditional EEG. Compared to EEG signals from normal children, epileptic discharges from clinical and subclinical childhood absence epilepsy seizures were positive and amplitude was significantly greater （P〈0.05）.CONCLUSION： Wavelet transformation was used to analyze EEG signals from childhood absence epilepsy to obtain multi-scale quantitative characteristics and phase average waveforms. Multi-scale wavelet coefficients of EEG signals correlated with childhood absence epilepsy seizure, and multi-scale waveforms prior to epilepsy seizure were similar to characteristics during the onset period. Compared to normal children, EEG signals during epilepsy seizure exhibited an opposed phase model.
基金Project supported by the National Natural Science Foundation of China(Grant No.61402368)Aerospace Support Fund,China(Grant No.2017-HT-XGD)Aerospace Science and Technology Innovation Foundation,China(Grant No.2017 ZD 53047)
文摘The high-frequency components in the traditional multi-scale transform method are approximately sparse, which can represent different information of the details. But in the low-frequency component, the coefficients around the zero value are very few, so we cannot sparsely represent low-frequency image information. The low-frequency component contains the main energy of the image and depicts the profile of the image. Direct fusion of the low-frequency component will not be conducive to obtain highly accurate fusion result. Therefore, this paper presents an infrared and visible image fusion method combining the multi-scale and top-hat transforms. On one hand, the new top-hat-transform can effectively extract the salient features of the low-frequency component. On the other hand, the multi-scale transform can extract highfrequency detailed information in multiple scales and from diverse directions. The combination of the two methods is conducive to the acquisition of more characteristics and more accurate fusion results. Among them, for the low-frequency component, a new type of top-hat transform is used to extract low-frequency features, and then different fusion rules are applied to fuse the low-frequency features and low-frequency background; for high-frequency components, the product of characteristics method is used to integrate the detailed information in high-frequency. Experimental results show that the proposed algorithm can obtain more detailed information and clearer infrared target fusion results than the traditional multiscale transform methods. Compared with the state-of-the-art fusion methods based on sparse representation, the proposed algorithm is simple and efficacious, and the time consumption is significantly reduced.
基金supported in part by the National Natural Science Foundation of China under Grant 62201201the Foundation of Henan Educational Committee under Grant 242102211042.
文摘Segmenting skin lesions is critical for early skin cancer detection.Existing CNN and Transformer-based methods face challenges such as high computational complexity and limited adaptability to variations in lesion sizes.To overcome these limitations,we introduce MSAMamba-UNet,a lightweight model that integrates two novel architectures:Multi-Scale Mamba(MSMamba)and Adaptive Dynamic Gating Block(ADGB).MSMamba utilizes multi-scale decomposition and a parallel hierarchical structure to enhance the delineation of irregular lesion boundaries and sensitivity to small targets.ADGB dynamically selects convolutional kernels with varying receptive fields based on input features,improving the model’s capacity to accommodate diverse lesion textures and scales.Additionally,we introduce a Mix Attention Fusion Block(MAF)to enhance shallow feature representation by integrating parallel channel and pixel attention mechanisms.Extensive evaluation of MSAMamba-UNet on the ISIC 2016,ISIC 2017,and ISIC 2018 datasets demonstrates competitive segmentation accuracy with only 0.056 M parameters and 0.069 GFLOPs.Our experiments revealed that MSAMamba-UNet achieved IoU scores of 85.53%,85.47%,and 82.22%,as well as DSC scores of 92.20%,92.17%,and 90.24%,respectively.These results underscore the lightweight design and effectiveness of MSAMamba-UNet.
基金Project supported by the National Basic Research Program of China(Grant Nos.2012CB720101 and 2012CB720103)the National Natural Science Foundation of China(Grant Nos.11272233,11332006,and 11411130150)
文摘When using a miniature single sensor boundary layer probe, the time sequences of the stream-wise velocity in the turbulent boundary layer （TBL） are measured by using a hot wire anemometer. Beneath the fully developed TBL, the wall pressure fluctuations are attained by a microphone mechanism with high spatial resolution. Analysis on the statistic and spectrum properties of velocity and wall pressure reveals the relationship between the wall pressure fluctuation and the energy-containing structure in the buffer layer of the TBL. Wavelet transform shows the multi-scale natures of coherent structures contained in both signals of velocity and pressure. The most intermittent wall pressure scale is associated with the coherent structure in the buffer layer. Meanwhile the most energetic scale of velocity fluctuation at y＋ = 14 provides a specific frequency f9 ≈ 147 Hz for wall actuating control with Ret = 996.
基金National Key Research and Development Program of China(Grant No.2017YFC1502102)National Natural Science Foundation of China(Grant No.42075148)+1 种基金Graduate Research and Innovation Projects of Jiangsu Province(Grant No.KYCX20_0910)the High-Performance Computing Center of Nanjing University of Information Science and Technology(NUIST).
文摘Use of data assimilation to initialize hydrometeors plays a vital role in numerical weather prediction(NWP).To directly analyze hydrometeors in data assimilation systems from cloud-sensitive observations,hydrometeor control variables are necessary.Common data assimilation systems theoretically require that the probability density functions(PDFs)of analysis,background,and observation errors should satisfy the Gaussian unbiased assumptions.In this study,a Gaussian transform method is proposed to transform hydrometeors to more Gaussian variables,which is modified from the Softmax function and renamed as Quasi-Softmax transform.The Quasi-Softmax transform method then is compared to the original hydrometeor mixing ratios and their logarithmic transform and Softmax transform.The spatial distribution,the non-Gaussian nature of the background errors,and the characteristics of the background errors of hydrometeors in each method are studied.Compared to the logarithmic and Softmax transform,the Quasi-Softmax method keeps the vertical distribution of the original hydrometeor mixing ratios to the greatest extent.The results of the D′Agostino test show that the hydrometeors transformed by the Quasi-Softmax method are more Gaussian when compared to the other methods.The Gaussian transform has been added to the control variable transform to estimate the background error covariances.Results show that the characteristics of the hydrometeor background errors are reasonable for the Quasi-Softmax method.The transformed hydrometeors using the Quasi-Softmax transform meet the Gaussian unbiased assumptions of the data assimilation system,and are promising control variables for data assimilation systems.
文摘Due to the lack of long-range association and spatial location information,fine details and accurate boundaries of complex clothing images cannot always be obtained by using the existing deep learning-based methods.This paper presents a convolutional structure with multi-scale fusion to optimize the step of clothing feature extraction and a self-attention module to capture long-range association information.The structure enables the self-attention mechanism to directly participate in the process of information exchange through the down-scaling projection operation of the multi-scale framework.In addition,the improved self-attention module introduces the extraction of 2-dimensional relative position information to make up for its lack of ability to extract spatial position features from clothing images.The experimental results based on the colorful fashion parsing dataset(CFPD)show that the proposed network structure achieves 53.68%mean intersection over union(mIoU)and has better performance on the clothing parsing task.
文摘Feature extraction of signals plays an important role in classification problems because of data dimension reduction property and potential improvement of a classification accuracy rate. Principal component analysis (PCA), wavelets transform or Fourier transform methods are often used for feature extraction. In this paper, we propose a multi-scale PCA, which combines discrete wavelet transform, and PCA for feature extraction of signals in both the spatial and temporal domains. Our study shows that the multi-scale PCA combined with the proposed new classification methods leads to high classification accuracy for the considered signals.
文摘Laser transformation hardening(LTH) is one of the laser surface modification processes.The surface hardening of rod-shaped carbon steel(SM45C) was performed by lathe-based laser composite processor with Gaussian-beam optical head.The LTH characteristics by dominant processes,longitudinal and depth directional hardness distributions and behaviors of phase transformation in hardened zones were examined.Especially,two concepts of circumferential speed and theoretical overlap rate were applied.When laser power increased or circumferential speed decreased,the surface hardening depth gradually increases due to the increased heat input.Moreover,the longitudinal hardness distribution particularly shows periodicity of repetitive increase and decrease,which results from tempering effect by overlap.Finally,the feasibility of laser transformation hardening is verified by using the beam with Gaussian intensity distribution.
基金National Natural Science Foundation of China(51779057,51709061,51509057)the Equipment Pre-Research Project(41412030201)the National 863 High Technology Development Plan Project(2011AA09A106)。
文摘To deal with the low location accuracy issue of existing underwater navigation technologies in autonomous underwater vehicles(AUVs),a distributed fusion algorithm which combines the model's analysis method with a multi-scale transformation method is proposed for integrated navigation system based on AUV.First,integrated navigation system theory and system error sources are introduced in details.Secondly,a navigation system's observation equation on the original scale is decomposed into different scales by the discrete wavelet transform method,and noise reduction is performed by setting the wavelet de-noising threshold.At last,the dynamic equation and observation equations are fused on different scales by the wavelet transformation and Kalman filter.The results show that the proposed algorithm has smaller navigation error and higher navigation accuracy.
基金supported in part by the National Natural Science Foundation of China(Grant No.62062003)Natural Science Foundation of Ningxia(Grant No.2023AAC03293).
文摘Computer-aided diagnosis of pneumonia based on deep learning is a research hotspot.However,there are some problems that the features of different sizes and different directions are not sufficient when extracting the features in lung X-ray images.A pneumonia classification model based on multi-scale directional feature enhancement MSD-Net is proposed in this paper.The main innovations are as follows:Firstly,the Multi-scale Residual Feature Extraction Module(MRFEM)is designed to effectively extract multi-scale features.The MRFEM uses dilated convolutions with different expansion rates to increase the receptive field and extract multi-scale features effectively.Secondly,the Multi-scale Directional Feature Perception Module(MDFPM)is designed,which uses a three-branch structure of different sizes convolution to transmit direction feature layer by layer,and focuses on the target region to enhance the feature information.Thirdly,the Axial Compression Former Module(ACFM)is designed to perform global calculations to enhance the perception ability of global features in different directions.To verify the effectiveness of the MSD-Net,comparative experiments and ablation experiments are carried out.In the COVID-19 RADIOGRAPHY DATABASE,the Accuracy,Recall,Precision,F1 Score,and Specificity of MSD-Net are 97.76%,95.57%,95.52%,95.52%,and 98.51%,respectively.In the chest X-ray dataset,the Accuracy,Recall,Precision,F1 Score and Specificity of MSD-Net are 97.78%,95.22%,96.49%,95.58%,and 98.11%,respectively.This model improves the accuracy of lung image recognition effectively and provides an important clinical reference to pneumonia Computer-Aided Diagnosis.
基金jointly sponsored by the Shenzhen Science and Technology Innovation Commission (Grant No. KCXFZ20201221173610028)the key program of the National Natural Science Foundation of China (Grant No. 42130605)
文摘In the traditional incremental analysis update(IAU)process,all analysis increments are treated as constant forcing in a model’s prognostic equations over a certain time window.This approach effectively reduces high-frequency oscillations introduced by data assimilation.However,as different scales of increments have unique evolutionary speeds and life histories in a numerical model,the traditional IAU scheme cannot fully meet the requirements of short-term forecasting for the damping of high-frequency noise and may even cause systematic drifts.Therefore,a multi-scale IAU scheme is proposed in this paper.Analysis increments were divided into different scale parts using a spatial filtering technique.For each scale increment,the optimal relaxation time in the IAU scheme was determined by the skill of the forecasting results.Finally,different scales of analysis increments were added to the model integration during their optimal relaxation time.The multi-scale IAU scheme can effectively reduce the noise and further improve the balance between large-scale and small-scale increments in the model initialization stage.To evaluate its performance,several numerical experiments were conducted to simulate the path and intensity of Typhoon Mangkhut(2018)and showed that:(1)the multi-scale IAU scheme had an obvious effect on noise control at the initial stage of data assimilation;(2)the optimal relaxation time for large-scale and small-scale increments was estimated as 6 h and 3 h,respectively;(3)the forecast performance of the multi-scale IAU scheme in the prediction of Typhoon Mangkhut(2018)was better than that of the traditional IAU scheme.The results demonstrate the superiority of the multi-scale IAU scheme.
