In pedestrian re-recognition,the traditional pedestrian re-recognition method will be affected by the changes of background,veil,clothing and so on,which will make the recognition effect decline.In order to reduce the...In pedestrian re-recognition,the traditional pedestrian re-recognition method will be affected by the changes of background,veil,clothing and so on,which will make the recognition effect decline.In order to reduce the impact of background,veil,clothing and other changes on the recognition effect,this paper proposes a pedestrian re-recognition method based on the cycle-consistent generative adversarial network and multifeature fusion.By comparing the measured distance between two pedestrians,pedestrian re-recognition is accomplished.Firstly,this paper uses Cycle GAN to transform and expand the data set,so as to reduce the influence of pedestrian posture changes as much as possible.The method consists of two branches:global feature extraction and local feature extraction.Then the global feature and local feature are fused.The fused features are used for comparison measurement learning,and the similarity scores are calculated to sort the samples.A large number of experimental results on large data sets CUHK03 and VIPER show that this new method reduces the influence of background,veil,clothing and other changes on the recognition effect.展开更多
The Internet of Things(IoT)is integral to modern infrastructure,enabling connectivity among a wide range of devices from home automation to industrial control systems.With the exponential increase in data generated by...The Internet of Things(IoT)is integral to modern infrastructure,enabling connectivity among a wide range of devices from home automation to industrial control systems.With the exponential increase in data generated by these interconnected devices,robust anomaly detection mechanisms are essential.Anomaly detection in this dynamic environment necessitates methods that can accurately distinguish between normal and anomalous behavior by learning intricate patterns.This paper presents a novel approach utilizing generative adversarial networks(GANs)for anomaly detection in IoT systems.However,optimizing GANs involves tuning hyper-parameters such as learning rate,batch size,and optimization algorithms,which can be challenging due to the non-convex nature of GAN loss functions.To address this,we propose a five-dimensional Gray wolf optimizer(5DGWO)to optimize GAN hyper-parameters.The 5DGWO introduces two new types of wolves:gamma(γ)for improved exploitation and convergence,and theta(θ)for enhanced exploration and escaping local minima.The proposed system framework comprises four key stages:1)preprocessing,2)generative model training,3)autoencoder(AE)training,and 4)predictive model training.The generative models are utilized to assist the AE training,and the final predictive models(including convolutional neural network(CNN),deep belief network(DBN),recurrent neural network(RNN),random forest(RF),and extreme gradient boosting(XGBoost))are trained using the generated data and AE-encoded features.We evaluated the system on three benchmark datasets:NSL-KDD,UNSW-NB15,and IoT-23.Experiments conducted on diverse IoT datasets show that our method outperforms existing anomaly detection strategies and significantly reduces false positives.The 5DGWO-GAN-CNNAE exhibits superior performance in various metrics,including accuracy,recall,precision,root mean square error(RMSE),and convergence trend.The proposed 5DGWO-GAN-CNNAE achieved the lowest RMSE values across the NSL-KDD,UNSW-NB15,and IoT-23 datasets,with values of 0.24,1.10,and 0.09,respectively.Additionally,it attained the highest accuracy,ranging from 94%to 100%.These results suggest a promising direction for future IoT security frameworks,offering a scalable and efficient solution to safeguard against evolving cyber threats.展开更多
The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper...The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper proposes an architecture that integrates speech prompts as input to image-generation Generative Adversarial Networks(GANs)model,leveraging Speech-to-Text translation along with the CLIP+VQGAN model.The proposed method involves translating speech prompts into text,which is then used by the Contrastive Language-Image Pretraining(CLIP)+Vector Quantized Generative Adversarial Network(VQGAN)model to generate images.This paper outlines the steps required to implement such a model and describes in detail the methods used for evaluating the model.The GAN model successfully generates artwork from descriptions using speech and text prompts.Experimental outcomes of synthesized images demonstrate that the proposed methodology can produce beautiful abstract visuals containing elements from the input prompts.The model achieved a Frechet Inception Distance(FID)score of 28.75,showcasing its capability to produce high-quality and diverse images.The proposed model can find numerous applications in educational,artistic,and design spaces due to its ability to generate images using speech and the distinct abstract artistry of the output images.This capability is demonstrated by giving the model out-of-the-box prompts to generate never-before-seen images with plausible realistic qualities.展开更多
Multifocal metalenses are of great concern in optical communications,optical imaging and micro-optics systems,but their design is extremely challenging.In recent years,deep learning methods have provided novel solutio...Multifocal metalenses are of great concern in optical communications,optical imaging and micro-optics systems,but their design is extremely challenging.In recent years,deep learning methods have provided novel solutions to the design of optical planar devices.Here,an approach is proposed to explore the use of generative adversarial networks(GANs)to realize the design of metalenses with different focusing positions at dual wavelengths.This approach includes a forward network and an inverse network,where the former predicts the optical response of meta-atoms and the latter generates structures that meet specific requirements.Compared to the traditional search method,the inverse network demonstrates higher precision and efficiency in designing a dual-wavelength bifocal metalens.The results will provide insights and methodologies for the design of tunable wavelength metalenses,while also highlighting the potential of deep learning in optical device design.展开更多
Full waveform inversion(FWI)has showed great potential in the detection of musculoskeletal disease.However,FWI is an ill-posed inverse problem and has a high requirement on the initial model during the imaging process...Full waveform inversion(FWI)has showed great potential in the detection of musculoskeletal disease.However,FWI is an ill-posed inverse problem and has a high requirement on the initial model during the imaging process.An inaccurate initial model may lead to local minima in the inversion and unexpected imaging results caused by cycle-skipping phenomenon.Deep learning methods have been applied in musculoskeletal imaging,but need a large amount of data for training.Inspired by work related to generative adversarial networks with physical informed constrain,we proposed a method named as bone ultrasound imaging with physics informed generative adversarial network(BUIPIGAN)to achieve unsupervised multi-parameter imaging for musculoskeletal tissues,focusing on speed of sound(SOS)and density.In the in-silico experiments using a ring array transducer,conventional FWI methods and BUIPIGAN were employed for multiparameter imaging of two musculoskeletal tissue models.The results were evaluated based on visual appearance,structural similarity index measure(SSIM),signal-to-noise ratio(SNR),and relative error(RE).For SOS imaging of the tibia–fibula model,the proposed BUIPIGAN achieved accurate SOS imaging with best performance.The specific quantitative metrics for SOS imaging were SSIM 0.9573,SNR 28.70 dB,and RE 5.78%.For the multi-parameter imaging of the tibia–fibula and human forearm,the BUIPIGAN successfully reconstructed SOS and density distributions with SSIM above 94%,SNR above 21 dB,and RE below 10%.The BUIPIGAN also showed robustness across various noise levels(i.e.,30 dB,10 dB).The results demonstrated that the proposed BUIPIGAN can achieve high-accuracy SOS and density imaging,proving its potential for applications in musculoskeletal ultrasound imaging.展开更多
The generation of high-quality,realistic face generation has emerged as a key field of research in computer vision.This paper proposes a robust approach that combines a Super-Resolution Generative Adversarial Network(...The generation of high-quality,realistic face generation has emerged as a key field of research in computer vision.This paper proposes a robust approach that combines a Super-Resolution Generative Adversarial Network(SRGAN)with a Pyramid Attention Module(PAM)to enhance the quality of deep face generation.The SRGAN framework is designed to improve the resolution of generated images,addressing common challenges such as blurriness and a lack of intricate details.The Pyramid Attention Module further complements the process by focusing on multi-scale feature extraction,enabling the network to capture finer details and complex facial features more effectively.The proposed method was trained and evaluated over 100 epochs on the CelebA dataset,demonstrating consistent improvements in image quality and a marked decrease in generator and discriminator losses,reflecting the model’s capacity to learn and synthesize high-quality images effectively,given adequate computational resources.Experimental outcome demonstrates that the SRGAN model with PAM module has outperformed,yielding an aggregate discriminator loss of 0.055 for real,0.043 for fake,and a generator loss of 10.58 after training for 100 epochs.The model has yielded an structural similarity index measure of 0.923,that has outperformed the other models that are considered in the current study for analysis.展开更多
Tropical cyclones(TCs)are complex and powerful weather systems,and accurately forecasting their path,structure,and intensity remains a critical focus and challenge in meteorological research.In this paper,we propose a...Tropical cyclones(TCs)are complex and powerful weather systems,and accurately forecasting their path,structure,and intensity remains a critical focus and challenge in meteorological research.In this paper,we propose an Attention Spatio-Temporal predictive Generative Adversarial Network(AST-GAN)model for predicting the temporal and spatial distribution of TCs.The model forecasts the spatial distribution of TC wind speeds for the next 15 hours at 3-hour intervals,emphasizing the cyclone's center,high wind-speed areas,and its asymmetric structure.To effectively capture spatiotemporal feature transfer at different time steps,we employ a channel attention mechanism for feature selection,enhancing model performance and reducing parameter redundancy.We utilized High-Resolution Weather Research and Forecasting(HWRF)data to train our model,allowing it to assimilate a wide range of TC motion patterns.The model is versatile and can be applied to various complex scenarios,such as multiple TCs moving simultaneously or TCs approaching landfall.Our proposed model demonstrates superior forecasting performance,achieving a root-mean-square error(RMSE)of 0.71 m s^(-1)for overall wind speed and 2.74 m s^(-1)for maximum wind speed when benchmarked against ground truth data from HWRF.Furthermore,the model underwent optimization and independent testing using ERA5reanalysis data,showcasing its stability and scalability.After fine-tuning on the ERA5 dataset,the model achieved an RMSE of 1.33 m s^(-1)for wind speed and 1.75 m s^(-1)for maximum wind speed.The AST-GAN model outperforms other state-of-the-art models in RMSE on both the HWRF and ERA5 datasets,maintaining its superior performance and demonstrating its effectiveness for spatiotemporal prediction of TCs.展开更多
This study addresses the pressing challenge of generating realistic strong ground motion data for simulating earthquakes,a crucial component in pre-earthquake risk assessments and post-earthquake disaster evaluations,...This study addresses the pressing challenge of generating realistic strong ground motion data for simulating earthquakes,a crucial component in pre-earthquake risk assessments and post-earthquake disaster evaluations,particularly suited for regions with limited seismic data.Herein,we report a generative adversarial network(GAN)framework capable of simulating strong ground motions under various environmental conditions using only a small set of real earthquake records.The constructed GAN model generates ground motions based on continuous physical variables such as source distance,site conditions,and magnitude,effectively capturing the complexity and diversity of ground motions under different scenarios.This capability allows the proposed model to approximate real seismic data,making it applicable to a wide range of engineering purposes.Using the Shandong Pingyuan earthquake as an example,a specialized dataset was constructed based on regional real ground motion records.The response spectrum at target locations was obtained through inverse distance-weighted interpolation of actual response spectra,followed by continuous wavelet transform to derive the ground motion time histories at these locations.Through iterative parameter adjustments,the constructed GAN model learned the probability distribution of strong-motion data for this event.The trained model generated three-component ground-motion time histories with clear P-wave and S-wave characteristics,accurately reflecting the non-stationary nature of seismic records.Statistical comparisons between synthetic and real response spectra,waveform envelopes,and peak ground acceleration show a high degree of similarity,underscoring the effectiveness of the model in replicating both the statistical and physical characteristics of real ground motions.These findings validate the feasibility of GANs for generating realistic earthquake data in data-scarce regions,providing a reliable approach for enriching regional ground motion databases.Additionally,the results suggest that GAN-based networks are a powerful tool for building predictive models in seismic hazard analysis.展开更多
In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and th...In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and the gradient penalty to achieve data augmentation.To lower the threshold for implementing the proposed method,transfer learning is further introduced.The W-DC-GAN-GP-TL framework is thereby formed.This framework is evaluated on 3 different publicly available datasets to judge the quality of generated data.Through visual comparisons and the examination of two visualization methods(probability density function(PDF)and principal component analysis(PCA)),it is demonstrated that the generated data is hard to distinguish from the real data.The application of generated data for training a battery state model using transfer learning is further evaluated.Specifically,Bi-GRU-based and Transformer-based methods are implemented on 2 separate datasets for estimating state of health(SOH)and state of charge(SOC),respectively.The results indicate that the proposed framework demonstrates satisfactory performance in different scenarios:for the data replacement scenario,where real data are removed and replaced with generated data,the state estimator accuracy decreases only slightly;for the data enhancement scenario,the estimator accuracy is further improved.The estimation accuracy of SOH and SOC is as low as 0.69%and 0.58%root mean square error(RMSE)after applying the proposed framework.This framework provides a reliable method for enriching battery measurement data.It is a generalized framework capable of generating a variety of time series data.展开更多
Background:Enhancing the quality of images from retinal,MRI and echocardiography imaging shows promise with SR-GANs for medical imaging use.Using these networks,it is possible to produce high-quality images even from ...Background:Enhancing the quality of images from retinal,MRI and echocardiography imaging shows promise with SR-GANs for medical imaging use.Using these networks,it is possible to produce high-quality images even from low-quality medical scans.Methods:To do this,SR-GANs make use of growth from low to high resolutions in two 2×stages,multiple sizes of filters and powerful loss functions.The medical super-resolution network and denoising SR-GAN focus on problems such as image noise and artifacts to improve a photo’s stability,ability to extract features and how it looks.Results:Assessment by numbers has found that using SR-GAN-based approaches leads to marked improvements such as increases in the PSNR by up to 4.85 dB and improvements in the SSIM by between 0.04 and 0.05.Such improvements are better than traditional super-resolution methods which help doctors achieve clear images of the mitral valve in cardiac ultrasonography.Conclusion:It is anticipated that applying SR-GANs in clinical tasks will increase the accuracy of diagnoses,ease the workload for patients and widen the application of super-resolution methods in various medical procedures.The results prove that SR-GANs improve the picture quality of echocardiograms used for diagnosing mitral valve problems.Having proven the model in research settings,future studies should try to apply it to real-world clinical cases,test for its use across a range of imaging devices and perfect the system to ensure it is efficient for use in medical settings.展开更多
Symmetric encryption algorithms learned by the previous proposed end-to-end adversarial network encryption communication systems are deterministic.With the same key and same plaintext,the deterministic algorithm will ...Symmetric encryption algorithms learned by the previous proposed end-to-end adversarial network encryption communication systems are deterministic.With the same key and same plaintext,the deterministic algorithm will lead to the same ciphertext.This means that the key in the deterministic encryption algorithm can only be used once,thus the encryption is not practical.To solve this problem,a nondeterministic symmetric encryption end-to-end communication system based on generative adversarial networks is proposed.We design a nonce-based adversarial neural network model,where a“nonce”standing for“number used only once”is passed to communication participants,and does not need to be secret.Moreover,we optimize the network structure through adding Batch Normalization(BN)to the CNNs(Convolutional Neural Networks),selecting the appropriate activation functions,and setting appropriate CNNs parameters.Results of experiments and analysis show that our system can achieve non-deterministic symmetric encryption,where Alice encrypting the same plaintext with the key twice will generate different ciphertexts,and Bob can decrypt all these different ciphertexts of the same plaintext to the correct plaintext.And our proposed system has fast convergence and the correct rate of decryption when the plaintext length is 256 or even longer.展开更多
The increasing demand for radioauthorized applications in the 6G era necessitates enhanced monitoring and management of radio resources,particularly for precise control over the electromagnetic environment.The radio m...The increasing demand for radioauthorized applications in the 6G era necessitates enhanced monitoring and management of radio resources,particularly for precise control over the electromagnetic environment.The radio map serves as a crucial tool for describing signal strength distribution within the current electromagnetic environment.However,most existing algorithms rely on sparse measurements of radio strength,disregarding the impact of building information.In this paper,we propose a spectrum cartography(SC)algorithm that eliminates the need for relying on sparse ground-based radio strength measurements by utilizing a satellite network to collect data on buildings and transmitters.Our algorithm leverages Pix2Pix Generative Adversarial Network(GAN)to construct accurate radio maps using transmitter information within real geographical environments.Finally,simulation results demonstrate that our algorithm exhibits superior accuracy compared to previously proposed methods.展开更多
This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the traini...This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the training dataset,and one solution is applied to improve the distribution of the training data by augmenting minority class samples using a deep convolutional generative adversarial network.Experi.mental results demonstrate that retraining the deep learning model with the newly generated dataset leads to a new fast radio burst classifier,which effectively reduces false positives caused by periodic wide-band impulsive radio frequency interference,thereby enhancing the performance of the search pipeline.展开更多
An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extra...An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extraction,which are commonly faced by rolling bearings and lead to low diagnostic accuracy.Initially,dual models of the Wasserstein deep convolutional generative adversarial network incorporating gradient penalty(1D-2DWDCGAN)are constructed to augment the original dataset.A self-adaptive loss threshold control training strategy is introduced,and establishing a self-adaptive balancing mechanism for stable model training.Subsequently,a diagnostic model based on multidimensional feature fusion is designed,wherein complex features from various dimensions are extracted,merging the original signal waveform features,structured features,and time-frequency features into a deep composite feature representation that encompasses multiple dimensions and scales;thus,efficient and accurate small sample fault diagnosis is facilitated.Finally,an experiment between the bearing fault dataset of CaseWestern ReserveUniversity and the fault simulation experimental platformdataset of this research group shows that this method effectively supplements the dataset and remarkably improves the diagnostic accuracy.The diagnostic accuracy after data augmentation reached 99.94%and 99.87%in two different experimental environments,respectively.In addition,robustness analysis is conducted on the diagnostic accuracy of the proposed method under different noise backgrounds,verifying its good generalization performance.展开更多
In this paper,a data-driven topology optimization(TO)method is proposed for the efficient design of three-dimensional heat transfer structures.The presented method is composed of four parts.Firstly,the three-dimension...In this paper,a data-driven topology optimization(TO)method is proposed for the efficient design of three-dimensional heat transfer structures.The presented method is composed of four parts.Firstly,the three-dimensional heat transfer topology optimization(HTTO)dataset,composed of both design parameters and the corresponding HTTO configuration,is established by the solid isotropic material with penalization(SIMP)method.Secondly,a high-performance surrogate model,named ResUNet-assisted generative adversarial nets(ResUNet-GAN),is developed by combining ReUNet and generative and adversarial nets(GAN).Thirdly,the same-resolution(SR)ResUNet-GAN is deployed to design three-dimensional heat transfer configurations by feeding design parameters.Finally,the finite element mesh of the optimized configuration is refined by the cross-resolution(CR)ResUNet-GAN to obtain near-optimal three-dimensional heat transfer configurations.Compared with conventional TO methods,the proposed method has two outstanding advantages:(1)the developed surrogate model establishes the end-to-end mapping from the design parameters to the three-dimensional configuration without any need for optimization iterations and finite element analysis;(2)both the SR ResUNet-GAN and the CR ResUNet-GAN can be employed individually or in combination to achieve each function,according to the needs of heat transfer structures.The data-driven method provides an efficient design framework for three-dimensional practical engineering problems.展开更多
Climate models are vital for understanding and projecting global climate change and its associated impacts.However,these models suffer from biases that limit their accuracy in historical simulations and the trustworth...Climate models are vital for understanding and projecting global climate change and its associated impacts.However,these models suffer from biases that limit their accuracy in historical simulations and the trustworthiness of future projections.Addressing these challenges requires addressing internal variability,hindering the direct alignment between model simulations and observations,and thwarting conventional supervised learning methods.Here,we employ an unsupervised Cycle-consistent Generative Adversarial Network(CycleGAN),to correct daily Sea Surface Temperature(SST)simulations from the Community Earth System Model 2(CESM2).Our results reveal that the CycleGAN not only corrects climatological biases but also improves the simulation of major dynamic modes including the El Niño-Southern Oscillation(ENSO)and the Indian Ocean Dipole mode,as well as SST extremes.Notably,it substantially corrects climatological SST biases,decreasing the globally averaged Root-Mean-Square Error(RMSE)by 58%.Intriguingly,the CycleGAN effectively addresses the well-known excessive westward bias in ENSO SST anomalies,a common issue in climate models that traditional methods,like quantile mapping,struggle to rectify.Additionally,it substantially improves the simulation of SST extremes,raising the pattern correlation coefficient(PCC)from 0.56 to 0.88 and lowering the RMSE from 0.5 to 0.32.This enhancement is attributed to better representations of interannual,intraseasonal,and synoptic scales variabilities.Our study offers a novel approach to correct global SST simulations and underscores its effectiveness across different time scales and primary dynamical modes.展开更多
Landslides are destructive natural disasters that cause catastrophic damage and loss of life worldwide.Accurately predicting landslide displacement enables effective early warning and risk management.However,the limit...Landslides are destructive natural disasters that cause catastrophic damage and loss of life worldwide.Accurately predicting landslide displacement enables effective early warning and risk management.However,the limited availability of on-site measurement data has been a substantial obstacle in developing data-driven models,such as state-of-the-art machine learning(ML)models.To address these challenges,this study proposes a data augmentation framework that uses generative adversarial networks(GANs),a recent advance in generative artificial intelligence(AI),to improve the accuracy of landslide displacement prediction.The framework provides effective data augmentation to enhance limited datasets.A recurrent GAN model,RGAN-LS,is proposed,specifically designed to generate realistic synthetic multivariate time series that mimics the characteristics of real landslide on-site measurement data.A customized moment-matching loss is incorporated in addition to the adversarial loss in GAN during the training of RGAN-LS to capture the temporal dynamics and correlations in real time series data.Then,the synthetic data generated by RGAN-LS is used to enhance the training of long short-term memory(LSTM)networks and particle swarm optimization-support vector machine(PSO-SVM)models for landslide displacement prediction tasks.Results on two landslides in the Three Gorges Reservoir(TGR)region show a significant improvement in LSTM model prediction performance when trained on augmented data.For instance,in the case of the Baishuihe landslide,the average root mean square error(RMSE)increases by 16.11%,and the mean absolute error(MAE)by 17.59%.More importantly,the model’s responsiveness during mutational stages is enhanced for early warning purposes.However,the results have shown that the static PSO-SVM model only sees marginal gains compared to recurrent models such as LSTM.Further analysis indicates that an optimal synthetic-to-real data ratio(50%on the illustration cases)maximizes the improvements.This also demonstrates the robustness and effectiveness of supplementing training data for dynamic models to obtain better results.By using the powerful generative AI approach,RGAN-LS can generate high-fidelity synthetic landslide data.This is critical for improving the performance of advanced ML models in predicting landslide displacement,particularly when there are limited training data.Additionally,this approach has the potential to expand the use of generative AI in geohazard risk management and other research areas.展开更多
Low-dose computed tomography(LDCT)denoising is an indispensable procedure in the medical imaging field,which not only improves image quality,but can mitigate the potential hazard to patients caused by routine doses.De...Low-dose computed tomography(LDCT)denoising is an indispensable procedure in the medical imaging field,which not only improves image quality,but can mitigate the potential hazard to patients caused by routine doses.Despite the improvement in performance of the cycle-consistent generative adversarial network(CycleGAN)due to the well-paired CT images shortage,there is still a need to further reduce image noise while retaining detailed features.Inspired by the residual encoder–decoder convolutional neural network(RED-CNN)and U-Net,we propose a novel unsupervised model using CycleGAN for LDCT imaging,which injects a two-sided network into selective kernel networks(SK-NET)to adaptively select features,and uses the patchGAN discriminator to generate CT images with more detail maintenance,aided by added perceptual loss.Based on patch-based training,the experimental results demonstrated that the proposed SKFCycleGAN outperforms competing methods in both a clinical dataset and the Mayo dataset.The main advantages of our method lie in noise suppression and edge preservation.展开更多
This study addresses challenges in fetal magnetic resonance imaging (MRI) related to motion artifacts, maternal respiration, and hardware limitations. To enhance MRI quality, we employ deep learning techniques, specif...This study addresses challenges in fetal magnetic resonance imaging (MRI) related to motion artifacts, maternal respiration, and hardware limitations. To enhance MRI quality, we employ deep learning techniques, specifically utilizing Cycle GAN. Synthetic pairs of images, simulating artifacts in fetal MRI, are generated to train the model. Our primary contribution is the use of Cycle GAN for fetal MRI restoration, augmented by artificially corrupted data. We compare three approaches (supervised Cycle GAN, Pix2Pix, and Mobile Unet) for artifact removal. Experimental results demonstrate that the proposed supervised Cycle GAN effectively removes artifacts while preserving image details, as validated through Structural Similarity Index Measure (SSIM) and normalized Mean Absolute Error (MAE). The method proves comparable to alternatives but avoids the generation of spurious regions, which is crucial for medical accuracy.展开更多
Mechanically cleaved two-dimensional materials are random in size and thickness.Recognizing atomically thin flakes by human experts is inefficient and unsuitable for scalable production.Deep learning algorithms have b...Mechanically cleaved two-dimensional materials are random in size and thickness.Recognizing atomically thin flakes by human experts is inefficient and unsuitable for scalable production.Deep learning algorithms have been adopted as an alternative,nevertheless a major challenge is a lack of sufficient actual training images.Here we report the generation of synthetic two-dimensional materials images using StyleGAN3 to complement the dataset.DeepLabv3Plus network is trained with the synthetic images which reduces overfitting and improves recognition accuracy to over 90%.A semi-supervisory technique for labeling images is introduced to reduce manual efforts.The sharper edges recognized by this method facilitate material stacking with precise edge alignment,which benefits exploring novel properties of layered-material devices that crucially depend on the interlayer twist-angle.This feasible and efficient method allows for the rapid and high-quality manufacturing of atomically thin materials and devices.展开更多
文摘In pedestrian re-recognition,the traditional pedestrian re-recognition method will be affected by the changes of background,veil,clothing and so on,which will make the recognition effect decline.In order to reduce the impact of background,veil,clothing and other changes on the recognition effect,this paper proposes a pedestrian re-recognition method based on the cycle-consistent generative adversarial network and multifeature fusion.By comparing the measured distance between two pedestrians,pedestrian re-recognition is accomplished.Firstly,this paper uses Cycle GAN to transform and expand the data set,so as to reduce the influence of pedestrian posture changes as much as possible.The method consists of two branches:global feature extraction and local feature extraction.Then the global feature and local feature are fused.The fused features are used for comparison measurement learning,and the similarity scores are calculated to sort the samples.A large number of experimental results on large data sets CUHK03 and VIPER show that this new method reduces the influence of background,veil,clothing and other changes on the recognition effect.
基金described in this paper has been developed with in the project PRESECREL(PID2021-124502OB-C43)。
文摘The Internet of Things(IoT)is integral to modern infrastructure,enabling connectivity among a wide range of devices from home automation to industrial control systems.With the exponential increase in data generated by these interconnected devices,robust anomaly detection mechanisms are essential.Anomaly detection in this dynamic environment necessitates methods that can accurately distinguish between normal and anomalous behavior by learning intricate patterns.This paper presents a novel approach utilizing generative adversarial networks(GANs)for anomaly detection in IoT systems.However,optimizing GANs involves tuning hyper-parameters such as learning rate,batch size,and optimization algorithms,which can be challenging due to the non-convex nature of GAN loss functions.To address this,we propose a five-dimensional Gray wolf optimizer(5DGWO)to optimize GAN hyper-parameters.The 5DGWO introduces two new types of wolves:gamma(γ)for improved exploitation and convergence,and theta(θ)for enhanced exploration and escaping local minima.The proposed system framework comprises four key stages:1)preprocessing,2)generative model training,3)autoencoder(AE)training,and 4)predictive model training.The generative models are utilized to assist the AE training,and the final predictive models(including convolutional neural network(CNN),deep belief network(DBN),recurrent neural network(RNN),random forest(RF),and extreme gradient boosting(XGBoost))are trained using the generated data and AE-encoded features.We evaluated the system on three benchmark datasets:NSL-KDD,UNSW-NB15,and IoT-23.Experiments conducted on diverse IoT datasets show that our method outperforms existing anomaly detection strategies and significantly reduces false positives.The 5DGWO-GAN-CNNAE exhibits superior performance in various metrics,including accuracy,recall,precision,root mean square error(RMSE),and convergence trend.The proposed 5DGWO-GAN-CNNAE achieved the lowest RMSE values across the NSL-KDD,UNSW-NB15,and IoT-23 datasets,with values of 0.24,1.10,and 0.09,respectively.Additionally,it attained the highest accuracy,ranging from 94%to 100%.These results suggest a promising direction for future IoT security frameworks,offering a scalable and efficient solution to safeguard against evolving cyber threats.
基金funded by the Centre for Advanced Modelling and Geospatial Information Systems(CAMGIS),Faculty of Engineering and IT,University of Technology SydneyMoreover,supported by the Researchers Supporting Project,King Saud University,Riyadh,Saudi Arabia,under Ongoing Research Funding(ORF-2025-14).
文摘The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper proposes an architecture that integrates speech prompts as input to image-generation Generative Adversarial Networks(GANs)model,leveraging Speech-to-Text translation along with the CLIP+VQGAN model.The proposed method involves translating speech prompts into text,which is then used by the Contrastive Language-Image Pretraining(CLIP)+Vector Quantized Generative Adversarial Network(VQGAN)model to generate images.This paper outlines the steps required to implement such a model and describes in detail the methods used for evaluating the model.The GAN model successfully generates artwork from descriptions using speech and text prompts.Experimental outcomes of synthesized images demonstrate that the proposed methodology can produce beautiful abstract visuals containing elements from the input prompts.The model achieved a Frechet Inception Distance(FID)score of 28.75,showcasing its capability to produce high-quality and diverse images.The proposed model can find numerous applications in educational,artistic,and design spaces due to its ability to generate images using speech and the distinct abstract artistry of the output images.This capability is demonstrated by giving the model out-of-the-box prompts to generate never-before-seen images with plausible realistic qualities.
基金National Natural Science Foundation of China(No.61975029)。
文摘Multifocal metalenses are of great concern in optical communications,optical imaging and micro-optics systems,but their design is extremely challenging.In recent years,deep learning methods have provided novel solutions to the design of optical planar devices.Here,an approach is proposed to explore the use of generative adversarial networks(GANs)to realize the design of metalenses with different focusing positions at dual wavelengths.This approach includes a forward network and an inverse network,where the former predicts the optical response of meta-atoms and the latter generates structures that meet specific requirements.Compared to the traditional search method,the inverse network demonstrates higher precision and efficiency in designing a dual-wavelength bifocal metalens.The results will provide insights and methodologies for the design of tunable wavelength metalenses,while also highlighting the potential of deep learning in optical device design.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12122403 and 12327807).
文摘Full waveform inversion(FWI)has showed great potential in the detection of musculoskeletal disease.However,FWI is an ill-posed inverse problem and has a high requirement on the initial model during the imaging process.An inaccurate initial model may lead to local minima in the inversion and unexpected imaging results caused by cycle-skipping phenomenon.Deep learning methods have been applied in musculoskeletal imaging,but need a large amount of data for training.Inspired by work related to generative adversarial networks with physical informed constrain,we proposed a method named as bone ultrasound imaging with physics informed generative adversarial network(BUIPIGAN)to achieve unsupervised multi-parameter imaging for musculoskeletal tissues,focusing on speed of sound(SOS)and density.In the in-silico experiments using a ring array transducer,conventional FWI methods and BUIPIGAN were employed for multiparameter imaging of two musculoskeletal tissue models.The results were evaluated based on visual appearance,structural similarity index measure(SSIM),signal-to-noise ratio(SNR),and relative error(RE).For SOS imaging of the tibia–fibula model,the proposed BUIPIGAN achieved accurate SOS imaging with best performance.The specific quantitative metrics for SOS imaging were SSIM 0.9573,SNR 28.70 dB,and RE 5.78%.For the multi-parameter imaging of the tibia–fibula and human forearm,the BUIPIGAN successfully reconstructed SOS and density distributions with SSIM above 94%,SNR above 21 dB,and RE below 10%.The BUIPIGAN also showed robustness across various noise levels(i.e.,30 dB,10 dB).The results demonstrated that the proposed BUIPIGAN can achieve high-accuracy SOS and density imaging,proving its potential for applications in musculoskeletal ultrasound imaging.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(*MSIT)(No.2018R1A5A7059549).
文摘The generation of high-quality,realistic face generation has emerged as a key field of research in computer vision.This paper proposes a robust approach that combines a Super-Resolution Generative Adversarial Network(SRGAN)with a Pyramid Attention Module(PAM)to enhance the quality of deep face generation.The SRGAN framework is designed to improve the resolution of generated images,addressing common challenges such as blurriness and a lack of intricate details.The Pyramid Attention Module further complements the process by focusing on multi-scale feature extraction,enabling the network to capture finer details and complex facial features more effectively.The proposed method was trained and evaluated over 100 epochs on the CelebA dataset,demonstrating consistent improvements in image quality and a marked decrease in generator and discriminator losses,reflecting the model’s capacity to learn and synthesize high-quality images effectively,given adequate computational resources.Experimental outcome demonstrates that the SRGAN model with PAM module has outperformed,yielding an aggregate discriminator loss of 0.055 for real,0.043 for fake,and a generator loss of 10.58 after training for 100 epochs.The model has yielded an structural similarity index measure of 0.923,that has outperformed the other models that are considered in the current study for analysis.
基金supported by the Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(NO.SML2021SP201)the National Natural Science Foundation of China(Grant No.42306200 and 42306216)+2 种基金the National Key Research and Development Program of China(Grant No.2023YFC3008100)the Innovation Group Project of the Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(Grant No.311021004)the Oceanic Interdisciplinary Program of Shanghai Jiao Tong University(Project No.SL2021ZD203)。
文摘Tropical cyclones(TCs)are complex and powerful weather systems,and accurately forecasting their path,structure,and intensity remains a critical focus and challenge in meteorological research.In this paper,we propose an Attention Spatio-Temporal predictive Generative Adversarial Network(AST-GAN)model for predicting the temporal and spatial distribution of TCs.The model forecasts the spatial distribution of TC wind speeds for the next 15 hours at 3-hour intervals,emphasizing the cyclone's center,high wind-speed areas,and its asymmetric structure.To effectively capture spatiotemporal feature transfer at different time steps,we employ a channel attention mechanism for feature selection,enhancing model performance and reducing parameter redundancy.We utilized High-Resolution Weather Research and Forecasting(HWRF)data to train our model,allowing it to assimilate a wide range of TC motion patterns.The model is versatile and can be applied to various complex scenarios,such as multiple TCs moving simultaneously or TCs approaching landfall.Our proposed model demonstrates superior forecasting performance,achieving a root-mean-square error(RMSE)of 0.71 m s^(-1)for overall wind speed and 2.74 m s^(-1)for maximum wind speed when benchmarked against ground truth data from HWRF.Furthermore,the model underwent optimization and independent testing using ERA5reanalysis data,showcasing its stability and scalability.After fine-tuning on the ERA5 dataset,the model achieved an RMSE of 1.33 m s^(-1)for wind speed and 1.75 m s^(-1)for maximum wind speed.The AST-GAN model outperforms other state-of-the-art models in RMSE on both the HWRF and ERA5 datasets,maintaining its superior performance and demonstrating its effectiveness for spatiotemporal prediction of TCs.
基金Funded by the National Key Research and Development Program(2022YFC3003502).
文摘This study addresses the pressing challenge of generating realistic strong ground motion data for simulating earthquakes,a crucial component in pre-earthquake risk assessments and post-earthquake disaster evaluations,particularly suited for regions with limited seismic data.Herein,we report a generative adversarial network(GAN)framework capable of simulating strong ground motions under various environmental conditions using only a small set of real earthquake records.The constructed GAN model generates ground motions based on continuous physical variables such as source distance,site conditions,and magnitude,effectively capturing the complexity and diversity of ground motions under different scenarios.This capability allows the proposed model to approximate real seismic data,making it applicable to a wide range of engineering purposes.Using the Shandong Pingyuan earthquake as an example,a specialized dataset was constructed based on regional real ground motion records.The response spectrum at target locations was obtained through inverse distance-weighted interpolation of actual response spectra,followed by continuous wavelet transform to derive the ground motion time histories at these locations.Through iterative parameter adjustments,the constructed GAN model learned the probability distribution of strong-motion data for this event.The trained model generated three-component ground-motion time histories with clear P-wave and S-wave characteristics,accurately reflecting the non-stationary nature of seismic records.Statistical comparisons between synthetic and real response spectra,waveform envelopes,and peak ground acceleration show a high degree of similarity,underscoring the effectiveness of the model in replicating both the statistical and physical characteristics of real ground motions.These findings validate the feasibility of GANs for generating realistic earthquake data in data-scarce regions,providing a reliable approach for enriching regional ground motion databases.Additionally,the results suggest that GAN-based networks are a powerful tool for building predictive models in seismic hazard analysis.
基金funded by the Bavarian State Ministry of Science,Research and Art(Grant number:H.2-F1116.WE/52/2)。
文摘In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and the gradient penalty to achieve data augmentation.To lower the threshold for implementing the proposed method,transfer learning is further introduced.The W-DC-GAN-GP-TL framework is thereby formed.This framework is evaluated on 3 different publicly available datasets to judge the quality of generated data.Through visual comparisons and the examination of two visualization methods(probability density function(PDF)and principal component analysis(PCA)),it is demonstrated that the generated data is hard to distinguish from the real data.The application of generated data for training a battery state model using transfer learning is further evaluated.Specifically,Bi-GRU-based and Transformer-based methods are implemented on 2 separate datasets for estimating state of health(SOH)and state of charge(SOC),respectively.The results indicate that the proposed framework demonstrates satisfactory performance in different scenarios:for the data replacement scenario,where real data are removed and replaced with generated data,the state estimator accuracy decreases only slightly;for the data enhancement scenario,the estimator accuracy is further improved.The estimation accuracy of SOH and SOC is as low as 0.69%and 0.58%root mean square error(RMSE)after applying the proposed framework.This framework provides a reliable method for enriching battery measurement data.It is a generalized framework capable of generating a variety of time series data.
文摘Background:Enhancing the quality of images from retinal,MRI and echocardiography imaging shows promise with SR-GANs for medical imaging use.Using these networks,it is possible to produce high-quality images even from low-quality medical scans.Methods:To do this,SR-GANs make use of growth from low to high resolutions in two 2×stages,multiple sizes of filters and powerful loss functions.The medical super-resolution network and denoising SR-GAN focus on problems such as image noise and artifacts to improve a photo’s stability,ability to extract features and how it looks.Results:Assessment by numbers has found that using SR-GAN-based approaches leads to marked improvements such as increases in the PSNR by up to 4.85 dB and improvements in the SSIM by between 0.04 and 0.05.Such improvements are better than traditional super-resolution methods which help doctors achieve clear images of the mitral valve in cardiac ultrasonography.Conclusion:It is anticipated that applying SR-GANs in clinical tasks will increase the accuracy of diagnoses,ease the workload for patients and widen the application of super-resolution methods in various medical procedures.The results prove that SR-GANs improve the picture quality of echocardiograms used for diagnosing mitral valve problems.Having proven the model in research settings,future studies should try to apply it to real-world clinical cases,test for its use across a range of imaging devices and perfect the system to ensure it is efficient for use in medical settings.
基金supported by The National Defense Innovation Project(No.ZZKY20222411)Natural Science Basic Research Plan in Shaanxi Province of China(No.2024JC-YBMS-546).
文摘Symmetric encryption algorithms learned by the previous proposed end-to-end adversarial network encryption communication systems are deterministic.With the same key and same plaintext,the deterministic algorithm will lead to the same ciphertext.This means that the key in the deterministic encryption algorithm can only be used once,thus the encryption is not practical.To solve this problem,a nondeterministic symmetric encryption end-to-end communication system based on generative adversarial networks is proposed.We design a nonce-based adversarial neural network model,where a“nonce”standing for“number used only once”is passed to communication participants,and does not need to be secret.Moreover,we optimize the network structure through adding Batch Normalization(BN)to the CNNs(Convolutional Neural Networks),selecting the appropriate activation functions,and setting appropriate CNNs parameters.Results of experiments and analysis show that our system can achieve non-deterministic symmetric encryption,where Alice encrypting the same plaintext with the key twice will generate different ciphertexts,and Bob can decrypt all these different ciphertexts of the same plaintext to the correct plaintext.And our proposed system has fast convergence and the correct rate of decryption when the plaintext length is 256 or even longer.
文摘The increasing demand for radioauthorized applications in the 6G era necessitates enhanced monitoring and management of radio resources,particularly for precise control over the electromagnetic environment.The radio map serves as a crucial tool for describing signal strength distribution within the current electromagnetic environment.However,most existing algorithms rely on sparse measurements of radio strength,disregarding the impact of building information.In this paper,we propose a spectrum cartography(SC)algorithm that eliminates the need for relying on sparse ground-based radio strength measurements by utilizing a satellite network to collect data on buildings and transmitters.Our algorithm leverages Pix2Pix Generative Adversarial Network(GAN)to construct accurate radio maps using transmitter information within real geographical environments.Finally,simulation results demonstrate that our algorithm exhibits superior accuracy compared to previously proposed methods.
基金supported by the Chinese Academy of Science"Light of West China"Program(2022-XBQNXZ-015)the National Natural Science Foundation of China(11903071)the Operation,Maintenance and Upgrading Fund for Astronomical Telescopes and Facility Instruments,budgeted from the Ministry of Finance of China and administered by the Chinese Academy of Sciences。
文摘This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the training dataset,and one solution is applied to improve the distribution of the training data by augmenting minority class samples using a deep convolutional generative adversarial network.Experi.mental results demonstrate that retraining the deep learning model with the newly generated dataset leads to a new fast radio burst classifier,which effectively reduces false positives caused by periodic wide-band impulsive radio frequency interference,thereby enhancing the performance of the search pipeline.
基金supported by the National Natural Science Foundation of China(Grant Nos.12272259 and 52005148).
文摘An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extraction,which are commonly faced by rolling bearings and lead to low diagnostic accuracy.Initially,dual models of the Wasserstein deep convolutional generative adversarial network incorporating gradient penalty(1D-2DWDCGAN)are constructed to augment the original dataset.A self-adaptive loss threshold control training strategy is introduced,and establishing a self-adaptive balancing mechanism for stable model training.Subsequently,a diagnostic model based on multidimensional feature fusion is designed,wherein complex features from various dimensions are extracted,merging the original signal waveform features,structured features,and time-frequency features into a deep composite feature representation that encompasses multiple dimensions and scales;thus,efficient and accurate small sample fault diagnosis is facilitated.Finally,an experiment between the bearing fault dataset of CaseWestern ReserveUniversity and the fault simulation experimental platformdataset of this research group shows that this method effectively supplements the dataset and remarkably improves the diagnostic accuracy.The diagnostic accuracy after data augmentation reached 99.94%and 99.87%in two different experimental environments,respectively.In addition,robustness analysis is conducted on the diagnostic accuracy of the proposed method under different noise backgrounds,verifying its good generalization performance.
基金supported by the National Natural Science Foundation of China(12472113,11872080)Natural Science Foundation of Beijing,China(3192005).
文摘In this paper,a data-driven topology optimization(TO)method is proposed for the efficient design of three-dimensional heat transfer structures.The presented method is composed of four parts.Firstly,the three-dimensional heat transfer topology optimization(HTTO)dataset,composed of both design parameters and the corresponding HTTO configuration,is established by the solid isotropic material with penalization(SIMP)method.Secondly,a high-performance surrogate model,named ResUNet-assisted generative adversarial nets(ResUNet-GAN),is developed by combining ReUNet and generative and adversarial nets(GAN).Thirdly,the same-resolution(SR)ResUNet-GAN is deployed to design three-dimensional heat transfer configurations by feeding design parameters.Finally,the finite element mesh of the optimized configuration is refined by the cross-resolution(CR)ResUNet-GAN to obtain near-optimal three-dimensional heat transfer configurations.Compared with conventional TO methods,the proposed method has two outstanding advantages:(1)the developed surrogate model establishes the end-to-end mapping from the design parameters to the three-dimensional configuration without any need for optimization iterations and finite element analysis;(2)both the SR ResUNet-GAN and the CR ResUNet-GAN can be employed individually or in combination to achieve each function,according to the needs of heat transfer structures.The data-driven method provides an efficient design framework for three-dimensional practical engineering problems.
基金supported by the National Natural Science Foundation of China(Grant Nos.42141019 and 42261144687)the Second Tibetan Plateau Scientific Expedition and Research(STEP)program(Grant No.2019QZKK0102)+4 种基金the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDB42010404)the National Natural Science Foundation of China(Grant No.42175049)the Guangdong Meteorological Service Science and Technology Research Project(Grant No.GRMC2021M01)the National Key Scientific and Technological Infrastructure project“Earth System Science Numerical Simulator Facility”(EarthLab)for computational support and Prof.Shiming XIANG for many useful discussionsNiklas BOERS acknowledges funding from the Volkswagen foundation.
文摘Climate models are vital for understanding and projecting global climate change and its associated impacts.However,these models suffer from biases that limit their accuracy in historical simulations and the trustworthiness of future projections.Addressing these challenges requires addressing internal variability,hindering the direct alignment between model simulations and observations,and thwarting conventional supervised learning methods.Here,we employ an unsupervised Cycle-consistent Generative Adversarial Network(CycleGAN),to correct daily Sea Surface Temperature(SST)simulations from the Community Earth System Model 2(CESM2).Our results reveal that the CycleGAN not only corrects climatological biases but also improves the simulation of major dynamic modes including the El Niño-Southern Oscillation(ENSO)and the Indian Ocean Dipole mode,as well as SST extremes.Notably,it substantially corrects climatological SST biases,decreasing the globally averaged Root-Mean-Square Error(RMSE)by 58%.Intriguingly,the CycleGAN effectively addresses the well-known excessive westward bias in ENSO SST anomalies,a common issue in climate models that traditional methods,like quantile mapping,struggle to rectify.Additionally,it substantially improves the simulation of SST extremes,raising the pattern correlation coefficient(PCC)from 0.56 to 0.88 and lowering the RMSE from 0.5 to 0.32.This enhancement is attributed to better representations of interannual,intraseasonal,and synoptic scales variabilities.Our study offers a novel approach to correct global SST simulations and underscores its effectiveness across different time scales and primary dynamical modes.
基金supported by the Natural Science Foundation of Jiangsu Province(Grant No.BK20220421)the State Key Program of the National Natural Science Foundation of China(Grant No.42230702)the National Natural Science Foundation of China(Grant No.82302352).
文摘Landslides are destructive natural disasters that cause catastrophic damage and loss of life worldwide.Accurately predicting landslide displacement enables effective early warning and risk management.However,the limited availability of on-site measurement data has been a substantial obstacle in developing data-driven models,such as state-of-the-art machine learning(ML)models.To address these challenges,this study proposes a data augmentation framework that uses generative adversarial networks(GANs),a recent advance in generative artificial intelligence(AI),to improve the accuracy of landslide displacement prediction.The framework provides effective data augmentation to enhance limited datasets.A recurrent GAN model,RGAN-LS,is proposed,specifically designed to generate realistic synthetic multivariate time series that mimics the characteristics of real landslide on-site measurement data.A customized moment-matching loss is incorporated in addition to the adversarial loss in GAN during the training of RGAN-LS to capture the temporal dynamics and correlations in real time series data.Then,the synthetic data generated by RGAN-LS is used to enhance the training of long short-term memory(LSTM)networks and particle swarm optimization-support vector machine(PSO-SVM)models for landslide displacement prediction tasks.Results on two landslides in the Three Gorges Reservoir(TGR)region show a significant improvement in LSTM model prediction performance when trained on augmented data.For instance,in the case of the Baishuihe landslide,the average root mean square error(RMSE)increases by 16.11%,and the mean absolute error(MAE)by 17.59%.More importantly,the model’s responsiveness during mutational stages is enhanced for early warning purposes.However,the results have shown that the static PSO-SVM model only sees marginal gains compared to recurrent models such as LSTM.Further analysis indicates that an optimal synthetic-to-real data ratio(50%on the illustration cases)maximizes the improvements.This also demonstrates the robustness and effectiveness of supplementing training data for dynamic models to obtain better results.By using the powerful generative AI approach,RGAN-LS can generate high-fidelity synthetic landslide data.This is critical for improving the performance of advanced ML models in predicting landslide displacement,particularly when there are limited training data.Additionally,this approach has the potential to expand the use of generative AI in geohazard risk management and other research areas.
基金funded by the National Natural Science Foundation of China(Grants No.61871277 and 61671312)in part by the Project of State Administration of Traditional Chinese Medicine of Sichuan(Grant No.2021MS012).
文摘Low-dose computed tomography(LDCT)denoising is an indispensable procedure in the medical imaging field,which not only improves image quality,but can mitigate the potential hazard to patients caused by routine doses.Despite the improvement in performance of the cycle-consistent generative adversarial network(CycleGAN)due to the well-paired CT images shortage,there is still a need to further reduce image noise while retaining detailed features.Inspired by the residual encoder–decoder convolutional neural network(RED-CNN)and U-Net,we propose a novel unsupervised model using CycleGAN for LDCT imaging,which injects a two-sided network into selective kernel networks(SK-NET)to adaptively select features,and uses the patchGAN discriminator to generate CT images with more detail maintenance,aided by added perceptual loss.Based on patch-based training,the experimental results demonstrated that the proposed SKFCycleGAN outperforms competing methods in both a clinical dataset and the Mayo dataset.The main advantages of our method lie in noise suppression and edge preservation.
文摘This study addresses challenges in fetal magnetic resonance imaging (MRI) related to motion artifacts, maternal respiration, and hardware limitations. To enhance MRI quality, we employ deep learning techniques, specifically utilizing Cycle GAN. Synthetic pairs of images, simulating artifacts in fetal MRI, are generated to train the model. Our primary contribution is the use of Cycle GAN for fetal MRI restoration, augmented by artificially corrupted data. We compare three approaches (supervised Cycle GAN, Pix2Pix, and Mobile Unet) for artifact removal. Experimental results demonstrate that the proposed supervised Cycle GAN effectively removes artifacts while preserving image details, as validated through Structural Similarity Index Measure (SSIM) and normalized Mean Absolute Error (MAE). The method proves comparable to alternatives but avoids the generation of spurious regions, which is crucial for medical accuracy.
基金Project supported by the National Key Research and Development Program of China(Grant No.2022YFB2803900)the National Natural Science Foundation of China(Grant Nos.61974075 and 61704121)+2 种基金the Natural Science Foundation of Tianjin Municipality(Grant Nos.22JCZDJC00460 and 19JCQNJC00700)Tianjin Municipal Education Commission(Grant No.2019KJ028)Fundamental Research Funds for the Central Universities(Grant No.22JCZDJC00460).
文摘Mechanically cleaved two-dimensional materials are random in size and thickness.Recognizing atomically thin flakes by human experts is inefficient and unsuitable for scalable production.Deep learning algorithms have been adopted as an alternative,nevertheless a major challenge is a lack of sufficient actual training images.Here we report the generation of synthetic two-dimensional materials images using StyleGAN3 to complement the dataset.DeepLabv3Plus network is trained with the synthetic images which reduces overfitting and improves recognition accuracy to over 90%.A semi-supervisory technique for labeling images is introduced to reduce manual efforts.The sharper edges recognized by this method facilitate material stacking with precise edge alignment,which benefits exploring novel properties of layered-material devices that crucially depend on the interlayer twist-angle.This feasible and efficient method allows for the rapid and high-quality manufacturing of atomically thin materials and devices.
