In LEO satellite communication networks,the number of satellites has increased sharply, the relative velocity of satellites is very fast, then electronic signal aliasing occurs from time to time. Those aliasing signal...In LEO satellite communication networks,the number of satellites has increased sharply, the relative velocity of satellites is very fast, then electronic signal aliasing occurs from time to time. Those aliasing signals make the receiving ability of the signal receiver worse, the signal processing ability weaker,and the anti-interference ability of the communication system lower. Aiming at the above problems, to save communication resources and improve communication efficiency, and considering the irregularity of interference signals, the underdetermined blind separation technology can effectively deal with the problem of interference sensing and signal reconstruction in this scenario. In order to improve the stability of source signal separation and the security of information transmission, a greedy optimization algorithm can be executed. At the same time, to improve network information transmission efficiency and prevent algorithms from getting trapped in local optima, delete low-energy points during each iteration process. Ultimately, simulation experiments validate that the algorithm presented in this paper enhances both the transmission efficiency of the network transmission system and the security of the communication system, achieving the process of interference sensing and signal reconstruction in the LEO satellite communication system.展开更多
A seepage-geomechanical coupled embedded fracture flow model has been established for multi-field coupled simulation in tight oil reservoirs,revealing the patterns of change in pressure field,seepage field,and stress ...A seepage-geomechanical coupled embedded fracture flow model has been established for multi-field coupled simulation in tight oil reservoirs,revealing the patterns of change in pressure field,seepage field,and stress field after long-term water injection in tight oil reservoirs.Based on this,a technique for enhanced oil recovery(EOR)combining multi-field reconstruction and combination of displacement and imbibition in tight oil reservoirs has been proposed.The study shows that after long-term water flooding for tight oil development,the pressure diffusion range is limited,making it difficult to establish an effective displacement system.The variation in geostress exhibits diversity,with the change in horizontal minimum principal stress being greater than that in horizontal maximum principal stress,and the variation around the injection wells being more significant than that around the production wells.The deflection of geostress direction around injection wells is also large.The technology for EOR through multi-field reconstruction and combination of displacement and imbibition employs water injection wells converted to production and large-scale fracturing techniques to restructure the artificial fracture network system.Through a full lifecycle energy replenishment method of pre-fracturing energy supplementation,energy increase during fracturing,well soaking for energy storage,and combination of displacement and imbibition,it effectively addresses the issue of easy channeling of the injection medium and difficult energy replenishment after large-scale fracturing.By intensifying the imbibition effect through the coordination of multiple wells,it reconstructs the combined system of displacement and imbibition under a complex fracture network,transitioning from avoiding fractures to utilizing them,thereby improving microscopic sweep and oil displacement efficiencies.Field application in Block Yuan 284 of the Huaqing Oilfield in the Ordos Basin has demonstrated that this technology increases the recovery factor by 12 percentage points,enabling large scale and efficient development of tight oil.展开更多
Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based...Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based on deep learning to achieve super resolution(SR)by utilizing deep singular-residual neural network(DSRNN)in training phase.Residuals are obtained from the difference between HR and LR images to generate LR-residual example pairs.Singular value decomposition(SVD)is applied to each LR-residual image pair to decompose into subbands of low and high frequency components.Later,DSRNN is trained on these subbands through input and output channels by optimizing the weights and biases of the network.With fewer layers in DSRNN,the influence of exploding gradients is reduced.This speeds up the learning process and also improves accuracy by using skip connections.The trained DSRNN parameters yield residuals to recover the HR subbands in the testing phase.Experimental analysis shows that the proposed method results in superior performance to existingmethods in terms of subjective quality.Extensive testing results on popular benchmark datasets such as set5,set14,and urban100 for a scaling factor of 4 show the effectiveness of the proposed method across different qualitative evaluation metrics.展开更多
At present,the acquisition of seismic data is developing toward high-precision and high-density methods.However,complex natural environments and cultural factors in many exploration areas cause difficulties in achievi...At present,the acquisition of seismic data is developing toward high-precision and high-density methods.However,complex natural environments and cultural factors in many exploration areas cause difficulties in achieving uniform and intensive acquisition,which makes complete seismic data collection impossible.Therefore,data reconstruction is required in the processing link to ensure imaging accuracy.Deep learning,as a new field in rapid development,presents clear advantages in feature extraction and modeling.In this study,the convolutional neural network deep learning algorithm is applied to seismic data reconstruction.Based on the convolutional neural network algorithm and combined with the characteristics of seismic data acquisition,two training strategies of supervised and unsupervised learning are designed to reconstruct sparse acquisition seismic records.First,a supervised learning strategy is proposed for labeled data,wherein the complete seismic data are segmented as the input of the training set and are randomly sampled before each training,thereby increasing the number of samples and the richness of features.Second,an unsupervised learning strategy based on large samples is proposed for unlabeled data,and the rolling segmentation method is used to update(pseudo)labels and training parameters in the training process.Through the reconstruction test of simulated and actual data,the deep learning algorithm based on a convolutional neural network shows better reconstruction quality and higher accuracy than compressed sensing based on Curvelet transform.展开更多
Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”k...Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”kernel size of the convolutional operator,for the spatially dense patterns,such as the generic face images,the performance of CNNs is limited.Here,we propose a“non-local”model,termed the Speckle-Transformer(SpT)UNet,for speckle feature extraction of generic face images.It is worth noting that the lightweight SpT UNet reveals a high efficiency and strong comparative performance with Pearson Correlation Coefficient(PCC),and structural similarity measure(SSIM)exceeding 0.989,and 0.950,respectively.展开更多
There are millimeter, micron and nanometer scales of pores and fractures in coal to describe different scales of coal pores and fissures communicating path and to quantitatively characterize their permeability. Such i...There are millimeter, micron and nanometer scales of pores and fractures in coal to describe different scales of coal pores and fissures communicating path and to quantitatively characterize their permeability. Such information provides an important basis for studying coalbed methane output mechanism. The pores and fissures in a large number of coal samples were observed and counted by scanning electron microscopy and optical microscopy. The probability distribution models of pore-fissure network were then established. Different scales of pore-fissures 2D network models were reconstructed by Monte Carlo method. The 2D seepage models were obtained through assignment zero method and using Matlab software. The effect of permeability on different scale pore-fractures network was obtained by two-dimensional seepage equation. Predicted permeability is compared with the measured ones. The results showed that the dominant order of different scale pore-fractures connected path from high to low is millimeter-sized fractures, seepage pores and micron-size fractures. The contribution of coal reservoir permeability from large to small is millimeter-size fractures, micron-size fractures and seepage pores. Different parameters in different scale pore-fractures are of different influence permeability.Reconstruction of different scale pore-fractures network can clearly display the connectivity of porefractures, which can provide a basis for selecting migration path and studying gas flow pattern.展开更多
In current research,a series of triaxial tests,which were employed to simulate three typical mining layouts(i.e.,top-coal caving,non-pillar mining and protected coal seam mining),were conducted on coal by using MTS815...In current research,a series of triaxial tests,which were employed to simulate three typical mining layouts(i.e.,top-coal caving,non-pillar mining and protected coal seam mining),were conducted on coal by using MTS815 Flex Test GT rock mechanics test system,and the fracture networks in the broken coal samples were qualitatively and quantitatively investigated by employing CT scanning and 3D reconstruction techniques.This work aimed at providing a detail description on the micro-structure and fractureconnectivity characteristics of rupture coal samples under different mining layouts.The results show that:(i)for protected coal seam mining layout,the coal specimens failure is in a compression-shear manner and oppositely,(ii)the tension-shear failure phenomenon is observed for top-coal caving and non-pillar mining layouts.By investigating the connectivity features of the generated fractures in the direction ofσ_1,under different mining layouts,it is found that the connectivity level of the fractures of the samples corresponding to non-pillar mining layout was the highest.展开更多
Phase space reconstruction is the first step of recognizing the chaotic time series.On the basis of differential entropy ratio method,the embedding dimension opt m and time delay t are optimal for the state space reco...Phase space reconstruction is the first step of recognizing the chaotic time series.On the basis of differential entropy ratio method,the embedding dimension opt m and time delay t are optimal for the state space reconstruction could be determined.But they are not the optimal parameters accepted for prediction.This study proposes an improved method based on the differential entropy ratio and Radial Basis Function(RBF)neural network to estimate the embedding dimension m and the time delay t,which have both optimal characteristics of the state space reconstruction and the prediction.Simulating experiments of Lorenz system and Doffing system show that the original phase space could be reconstructed from the time series effectively,and both the prediction accuracy and prediction length are improved greatly.展开更多
The aim is to reconstruct a complete and detailed clothed human from a single-view input.Implicit function is suitable for this task because it represents fine shape details and varied topology.Current methods,however...The aim is to reconstruct a complete and detailed clothed human from a single-view input.Implicit function is suitable for this task because it represents fine shape details and varied topology.Current methods,however,often suffer from artefacts such as broken or disembodied body parts,missing details,or depth ambiguity due to the ambiguity and complexity of human articulation.The main issue observed by the authors is structureagnostic.To address these problems,the authors fully utilise the skinned multi-person linear(SMPL)model and propose a method using the Skeleton-aware Implicit Function(SIF).To alleviate the broken or disembodied body parts,the proposed skeleton-aware structure prior makes the skeleton awareness into an implicit function,which consists of a bone-guided sampling strategy and a skeleton-relative encoding strategy.To deal with the missing details and depth ambiguity problems,the authors’body-guided pixel-aligned feature exploits the SMPL to enhance 2D normal and depth semantic features,and the proposed feature aggregation uses the extra geometry-aware prior to enabling a more plausible merging with less noisy geometry.Additionally,SIF is also adapted to the RGB-D input,and experimental results show that SIF outperforms the state-of-the-arts methods on challenging datasets from Twindom and Thuman3.0.展开更多
In this paper,we investigate the secrecy outage performance for the two-way integrated satellite unmanned aerial vehicle relay networks with hardware impairments.Particularly,the closed-form expression for the secrecy...In this paper,we investigate the secrecy outage performance for the two-way integrated satellite unmanned aerial vehicle relay networks with hardware impairments.Particularly,the closed-form expression for the secrecy outage probability is obtained.Moreover,to get more information on the secrecy outage probability in a high signalto-noise regime,the asymptotic analysis along with the secrecy diversity order and secrecy coding gain for the secrecy outage probability are also further obtained,which presents a fast method to evaluate the impact of system parameters and hardware impairments on the considered network.Finally,Monte Carlo simulation results are provided to show the efficiency of the theoretical analysis.展开更多
The state reconstruction problem is addressed for complex dynamical networks coupled with states and outputs respectively, in a noisy transmission channel. By using Lyapunov stability theory and H∞ performance, two s...The state reconstruction problem is addressed for complex dynamical networks coupled with states and outputs respectively, in a noisy transmission channel. By using Lyapunov stability theory and H∞ performance, two schemes of state reconstruction are proposed for the complex dynamical networks with the nodes coupled by states and outputs respectively, and the estimation errors are convergent to zeros with H∞ performance index. A numerical simulation demonstrates the effectiveness of the proposed observers.展开更多
The KT-II layer in the Zananor Oilfield,Caspian Basin,Kazakhstan,contains carbonate reservoirs of various types.The complex pore structure of the reservoirs have made it difficult to identify watered-out zones with tr...The KT-II layer in the Zananor Oilfield,Caspian Basin,Kazakhstan,contains carbonate reservoirs of various types.The complex pore structure of the reservoirs have made it difficult to identify watered-out zones with traditional logging interpretation methods.This study classifies the reservoirs on the basis of core analysis and establishes an identification model for watered-out layers in the field to effectively improve the interpretation accuracy.Thin section analysis shows that there are three types of pores in the reservoirs,i.e.,the matrix pore,fracture and dissolution vug.A triple porosity model is used to calculate the porosities of the reservoirs and the results are combined with core analysis to classify the reservoirs into the fractured,matrix pore,fracture-pore as well as composite types.A classification standard is also proposed.There are differences in resistivity logging responses from the reservoirs of different types before and after watering-out.The preewatering-out resistivities are reconstructed using generalized neural network for different types of reservoirs.The watered-out layers can be effectively identified according to the difference in resistivity curves before and after watering-out.The results show that the watered-out layers identified with the method are consistent with measured data,thus serving as a reference for the evaluation of watered-out layers in the study area.展开更多
In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reco...In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reconstructed by traditional associative imaging methods. Unlike traditional ghost imaging to reconstruct objects from bucket signals, our proposed method can use simple objects (such as EMNIST) as a training set for GAN, and then recognize objects (such as faces) of completely different complexity than the training set. We use traditional ghost imaging and neural network to reconstruct target objects respectively. According to the research results in this paper, the method based on neural network can reconstruct complex objects very well, but the method based on traditional ghost imaging cannot reconstruct complex objects. The research scheme in this paper is of great significance for the reconstruction of complex object-related imaging under low sampling conditions.展开更多
Proposed a new method to disclose the complicated non-linearity structure of the water-resource system, introducing chaos theory into the hydrology and water resources field, and combined with the chaos theory and art...Proposed a new method to disclose the complicated non-linearity structure of the water-resource system, introducing chaos theory into the hydrology and water resources field, and combined with the chaos theory and artificial neural networks. Training data construction and networks structure were determined by the phase space reconstruction, and establishing nonlinear relationship of phase points with neural networks, the forecasting model of the resource quantity of the surface water was brought forward. The keystone of the way and the detailed arithmetic of the network training were given. The example shows that the model has highly forecasting precision.展开更多
A new method of super-resolution image reconstruction is proposed, which uses a three-step-training error backpropagation neural network (BPNN) to realize the super-resolution reconstruction (SRR) of satellite ima...A new method of super-resolution image reconstruction is proposed, which uses a three-step-training error backpropagation neural network (BPNN) to realize the super-resolution reconstruction (SRR) of satellite image. The method is based on BPNN. First, three groups learning samples with different resolutions are obtained according to image observation model, and then vector mappings are respectively used to those three group learning samples to speed up the convergence of BPNN, at last, three times consecutive training are carried on the BPNN. Training samples used in each step are of higher resolution than those used in the previous steps, so the increasing weights store a great amount of information for SRR, and network performance and generalization ability are improved greatly. Simulation and generalization tests are carried on the well-trained three-step-training NN respectively, and the reconstruction results with higher resolution images verify the effectiveness and validity of this method.展开更多
As digital image techniques have been widely used, the requirements for high-resolution images become increasingly stringent. Traditional single-frame interpolation techniques cannot add new high frequency information...As digital image techniques have been widely used, the requirements for high-resolution images become increasingly stringent. Traditional single-frame interpolation techniques cannot add new high frequency information to the expanded images, and cannot improve resolution in deed. Multiframe-based techniques are effective ways for high-resolution image reconstruction, but their computation complexities and the difficulties in achieving image sequences limit their applications. An original method using an artificial neural network is proposed in this paper. Using the inherent merits in neural network, we can establish the mapping between high frequency components in low-resolution images and high-resolution images. Example applications and their results demonstrated the images reconstructed by our method are aesthetically and quantitatively (using the criteria of MSE and MAE) superior to the images acquired by common methods. Even for infrared images this method can give satisfactory results with high definition. In addition, a single-layer linear neural network is used in this paper, the computational complexity is very low, and this method can be realized in real time.展开更多
Background:Breast reconstruction is an effective technique to rebuild the appearance of the breasts in patients after mastectomy and improves the prognosis.The current study aimed to compare and analyze willingness fo...Background:Breast reconstruction is an effective technique to rebuild the appearance of the breasts in patients after mastectomy and improves the prognosis.The current study aimed to compare and analyze willingness for breast reconstruction after breast cancer between populations in China and the United States,from the perspective of social concern,using big data analysis.We also aimed to explore factors affecting surgical selection and to identify methods that can improve social cognition and acceptance of breast reconstruction.Methods:Using Baidu and Google,two representative Internet search engines in China and the United States as research tools,and using big data search volume as the benchmark,we compared and analyzed breast reconstruction willingness and attention characteristics between Chinese and American people,based on search heat,geographical distribution,age and sex,keyword distribution,ethnic group,and social development degree.Results:In both the long-term and short-term,Chinese people paid more attention towards searching about breast cancer,but less attention to breast reconstruction after breast cancer surgery.However,in both the short-term and long-term,people from the United States paid more attention towards breast cancer and breast reconstruction with the help of the Internet,showing a synchronous change relationship.There was a large regional difference in the search volume for breast cancer among the Chinese population,while no significant regional differences were noted in the search volume for breast cancer in the United States.However,a large regional difference was observed in the search volume for breast reconstruction between the two countries;people in the coastal and economically developed areas paid more attention to it.Most people who paid attention to breast reconstruction in China were women aged 20–39 years,while the attention among men was low.Search keywords were also limited to breast cancer-related information.However,between Asians and European Americans,Americans paid more attention to breast cancer and were affected by regional development,religious beliefs,and health facilities.Conclusion:Attention towards breast reconstruction after breast cancer was lower in the Chinese population than in the American population,and this difference was closely related to the level of regional development.There is insufficient information on breast reconstruction after breast cancer in recent Internet media.In addition to strengthening communication in clinics,media education is important to improve the cognitive level and social awareness of patients and their families,which is conducive to breast reconstruction.展开更多
In robot-assisted surgery projects,researchers should be able to make fast 3D reconstruction. Usually 2D images acquired with common diagnostic equipments such as UT, CT and MRI are not enough and complete for an accu...In robot-assisted surgery projects,researchers should be able to make fast 3D reconstruction. Usually 2D images acquired with common diagnostic equipments such as UT, CT and MRI are not enough and complete for an accurate 3D reconstruction. There are some interpolation methods for approximating non value voxels which consume large execution time. A novel algorithm is introduced based on generalized regression neural network (GRNN) which can interpolate unknown voxles fast and reliable. The GRNN interpolation is used to produce new 2D images between each two succeeding ultrasonic images. It is shown that the composition of GRNN with image distance transformation can produce higher quality 3D shapes. The results of this method are compared with other interpolation methods practically. It shows this method can decrease overall time consumption on online 3D reconstruction.展开更多
Super-resolution reconstruction in medical imaging has become more demanding due to the necessity of obtaining high-quality images with minimal radiation dose,such as in low-field magnetic resonance imaging(MRI).Howev...Super-resolution reconstruction in medical imaging has become more demanding due to the necessity of obtaining high-quality images with minimal radiation dose,such as in low-field magnetic resonance imaging(MRI).However,image super-resolution reconstruction remains a difficult task because of the complexity and high textual requirements for diagnosis purpose.In this paper,we offer a deep learning based strategy for reconstructing medical images from low resolutions utilizing Transformer and generative adversarial networks(T-GANs).The integrated system can extract more precise texture information and focus more on important locations through global image matching after successfully inserting Transformer into the generative adversarial network for picture reconstruction.Furthermore,we weighted the combination of content loss,adversarial loss,and adversarial feature loss as the final multi-task loss function during the training of our proposed model T-GAN.In comparison to established measures like peak signal-to-noise ratio(PSNR)and structural similarity index measure(SSIM),our suggested T-GAN achieves optimal performance and recovers more texture features in super-resolution reconstruction of MRI scanned images of the knees and belly.展开更多
基金supported by National Natural Science Foundation of China (62171390)Central Universities of Southwest Minzu University (ZYN2022032,2023NYXXS034)the State Scholarship Fund of the China Scholarship Council (NO.202008510081)。
文摘In LEO satellite communication networks,the number of satellites has increased sharply, the relative velocity of satellites is very fast, then electronic signal aliasing occurs from time to time. Those aliasing signals make the receiving ability of the signal receiver worse, the signal processing ability weaker,and the anti-interference ability of the communication system lower. Aiming at the above problems, to save communication resources and improve communication efficiency, and considering the irregularity of interference signals, the underdetermined blind separation technology can effectively deal with the problem of interference sensing and signal reconstruction in this scenario. In order to improve the stability of source signal separation and the security of information transmission, a greedy optimization algorithm can be executed. At the same time, to improve network information transmission efficiency and prevent algorithms from getting trapped in local optima, delete low-energy points during each iteration process. Ultimately, simulation experiments validate that the algorithm presented in this paper enhances both the transmission efficiency of the network transmission system and the security of the communication system, achieving the process of interference sensing and signal reconstruction in the LEO satellite communication system.
基金Supported by the Joint Fund Project of the National Natural Science Foundation of China(U22B2075).
文摘A seepage-geomechanical coupled embedded fracture flow model has been established for multi-field coupled simulation in tight oil reservoirs,revealing the patterns of change in pressure field,seepage field,and stress field after long-term water injection in tight oil reservoirs.Based on this,a technique for enhanced oil recovery(EOR)combining multi-field reconstruction and combination of displacement and imbibition in tight oil reservoirs has been proposed.The study shows that after long-term water flooding for tight oil development,the pressure diffusion range is limited,making it difficult to establish an effective displacement system.The variation in geostress exhibits diversity,with the change in horizontal minimum principal stress being greater than that in horizontal maximum principal stress,and the variation around the injection wells being more significant than that around the production wells.The deflection of geostress direction around injection wells is also large.The technology for EOR through multi-field reconstruction and combination of displacement and imbibition employs water injection wells converted to production and large-scale fracturing techniques to restructure the artificial fracture network system.Through a full lifecycle energy replenishment method of pre-fracturing energy supplementation,energy increase during fracturing,well soaking for energy storage,and combination of displacement and imbibition,it effectively addresses the issue of easy channeling of the injection medium and difficult energy replenishment after large-scale fracturing.By intensifying the imbibition effect through the coordination of multiple wells,it reconstructs the combined system of displacement and imbibition under a complex fracture network,transitioning from avoiding fractures to utilizing them,thereby improving microscopic sweep and oil displacement efficiencies.Field application in Block Yuan 284 of the Huaqing Oilfield in the Ordos Basin has demonstrated that this technology increases the recovery factor by 12 percentage points,enabling large scale and efficient development of tight oil.
文摘Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based on deep learning to achieve super resolution(SR)by utilizing deep singular-residual neural network(DSRNN)in training phase.Residuals are obtained from the difference between HR and LR images to generate LR-residual example pairs.Singular value decomposition(SVD)is applied to each LR-residual image pair to decompose into subbands of low and high frequency components.Later,DSRNN is trained on these subbands through input and output channels by optimizing the weights and biases of the network.With fewer layers in DSRNN,the influence of exploding gradients is reduced.This speeds up the learning process and also improves accuracy by using skip connections.The trained DSRNN parameters yield residuals to recover the HR subbands in the testing phase.Experimental analysis shows that the proposed method results in superior performance to existingmethods in terms of subjective quality.Extensive testing results on popular benchmark datasets such as set5,set14,and urban100 for a scaling factor of 4 show the effectiveness of the proposed method across different qualitative evaluation metrics.
基金This study was supported by the National Natural Science Foundation of China under the project‘Research on the Dynamic Location of Receiver Points and Wave Field Separation Technology Based on Deep Learning in OBN Seismic Exploration’(No.42074140).
文摘At present,the acquisition of seismic data is developing toward high-precision and high-density methods.However,complex natural environments and cultural factors in many exploration areas cause difficulties in achieving uniform and intensive acquisition,which makes complete seismic data collection impossible.Therefore,data reconstruction is required in the processing link to ensure imaging accuracy.Deep learning,as a new field in rapid development,presents clear advantages in feature extraction and modeling.In this study,the convolutional neural network deep learning algorithm is applied to seismic data reconstruction.Based on the convolutional neural network algorithm and combined with the characteristics of seismic data acquisition,two training strategies of supervised and unsupervised learning are designed to reconstruct sparse acquisition seismic records.First,a supervised learning strategy is proposed for labeled data,wherein the complete seismic data are segmented as the input of the training set and are randomly sampled before each training,thereby increasing the number of samples and the richness of features.Second,an unsupervised learning strategy based on large samples is proposed for unlabeled data,and the rolling segmentation method is used to update(pseudo)labels and training parameters in the training process.Through the reconstruction test of simulated and actual data,the deep learning algorithm based on a convolutional neural network shows better reconstruction quality and higher accuracy than compressed sensing based on Curvelet transform.
基金funding support from the Science and Technology Commission of Shanghai Municipality(Grant No.21DZ1100500)the Shanghai Frontiers Science Center Program(2021-2025 No.20)+2 种基金the Zhangjiang National Innovation Demonstration Zone(Grant No.ZJ2019ZD-005)supported by a fellowship from the China Postdoctoral Science Foundation(2020M671169)the International Postdoctoral Exchange Program from the Administrative Committee of Post-Doctoral Researchers of China([2020]33)。
文摘Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”kernel size of the convolutional operator,for the spatially dense patterns,such as the generic face images,the performance of CNNs is limited.Here,we propose a“non-local”model,termed the Speckle-Transformer(SpT)UNet,for speckle feature extraction of generic face images.It is worth noting that the lightweight SpT UNet reveals a high efficiency and strong comparative performance with Pearson Correlation Coefficient(PCC),and structural similarity measure(SSIM)exceeding 0.989,and 0.950,respectively.
基金in major projects of Henan Province University Science and Technology Innovation Talent Support Program of China (No. 15HASTIT050)Funding Scheme for Henan Province the Young Key Teachers (No. 2013GGJS-049) of ChinaScience and Technology Department of Henan Province of China (No. 142102210050)
文摘There are millimeter, micron and nanometer scales of pores and fractures in coal to describe different scales of coal pores and fissures communicating path and to quantitatively characterize their permeability. Such information provides an important basis for studying coalbed methane output mechanism. The pores and fissures in a large number of coal samples were observed and counted by scanning electron microscopy and optical microscopy. The probability distribution models of pore-fissure network were then established. Different scales of pore-fissures 2D network models were reconstructed by Monte Carlo method. The 2D seepage models were obtained through assignment zero method and using Matlab software. The effect of permeability on different scale pore-fractures network was obtained by two-dimensional seepage equation. Predicted permeability is compared with the measured ones. The results showed that the dominant order of different scale pore-fractures connected path from high to low is millimeter-sized fractures, seepage pores and micron-size fractures. The contribution of coal reservoir permeability from large to small is millimeter-size fractures, micron-size fractures and seepage pores. Different parameters in different scale pore-fractures are of different influence permeability.Reconstruction of different scale pore-fractures network can clearly display the connectivity of porefractures, which can provide a basis for selecting migration path and studying gas flow pattern.
基金financially supported by the Major State Fundamental Research Project of China(Nos.2011CB201201and2010CB226802)the National Natural Science Foundation of China(No.51204113)the Youth Science and Technology Fund of Sichuan Province(No.2012JQ0031)
文摘In current research,a series of triaxial tests,which were employed to simulate three typical mining layouts(i.e.,top-coal caving,non-pillar mining and protected coal seam mining),were conducted on coal by using MTS815 Flex Test GT rock mechanics test system,and the fracture networks in the broken coal samples were qualitatively and quantitatively investigated by employing CT scanning and 3D reconstruction techniques.This work aimed at providing a detail description on the micro-structure and fractureconnectivity characteristics of rupture coal samples under different mining layouts.The results show that:(i)for protected coal seam mining layout,the coal specimens failure is in a compression-shear manner and oppositely,(ii)the tension-shear failure phenomenon is observed for top-coal caving and non-pillar mining layouts.By investigating the connectivity features of the generated fractures in the direction ofσ_1,under different mining layouts,it is found that the connectivity level of the fractures of the samples corresponding to non-pillar mining layout was the highest.
基金Supported by the Key Program of National Natural Science Foundation of China(Nos.61077071,51075349)Program of National Natural Science Foundation of Hebei Province(Nos.F2011203207,F2010001312)
文摘Phase space reconstruction is the first step of recognizing the chaotic time series.On the basis of differential entropy ratio method,the embedding dimension opt m and time delay t are optimal for the state space reconstruction could be determined.But they are not the optimal parameters accepted for prediction.This study proposes an improved method based on the differential entropy ratio and Radial Basis Function(RBF)neural network to estimate the embedding dimension m and the time delay t,which have both optimal characteristics of the state space reconstruction and the prediction.Simulating experiments of Lorenz system and Doffing system show that the original phase space could be reconstructed from the time series effectively,and both the prediction accuracy and prediction length are improved greatly.
基金National Key R&D Program of China,Grant/Award Number:2022YFF0901902。
文摘The aim is to reconstruct a complete and detailed clothed human from a single-view input.Implicit function is suitable for this task because it represents fine shape details and varied topology.Current methods,however,often suffer from artefacts such as broken or disembodied body parts,missing details,or depth ambiguity due to the ambiguity and complexity of human articulation.The main issue observed by the authors is structureagnostic.To address these problems,the authors fully utilise the skinned multi-person linear(SMPL)model and propose a method using the Skeleton-aware Implicit Function(SIF).To alleviate the broken or disembodied body parts,the proposed skeleton-aware structure prior makes the skeleton awareness into an implicit function,which consists of a bone-guided sampling strategy and a skeleton-relative encoding strategy.To deal with the missing details and depth ambiguity problems,the authors’body-guided pixel-aligned feature exploits the SMPL to enhance 2D normal and depth semantic features,and the proposed feature aggregation uses the extra geometry-aware prior to enabling a more plausible merging with less noisy geometry.Additionally,SIF is also adapted to the RGB-D input,and experimental results show that SIF outperforms the state-of-the-arts methods on challenging datasets from Twindom and Thuman3.0.
基金supported by the Natural Science Foundation of China under Grant No.62001517.
文摘In this paper,we investigate the secrecy outage performance for the two-way integrated satellite unmanned aerial vehicle relay networks with hardware impairments.Particularly,the closed-form expression for the secrecy outage probability is obtained.Moreover,to get more information on the secrecy outage probability in a high signalto-noise regime,the asymptotic analysis along with the secrecy diversity order and secrecy coding gain for the secrecy outage probability are also further obtained,which presents a fast method to evaluate the impact of system parameters and hardware impairments on the considered network.Finally,Monte Carlo simulation results are provided to show the efficiency of the theoretical analysis.
文摘The state reconstruction problem is addressed for complex dynamical networks coupled with states and outputs respectively, in a noisy transmission channel. By using Lyapunov stability theory and H∞ performance, two schemes of state reconstruction are proposed for the complex dynamical networks with the nodes coupled by states and outputs respectively, and the estimation errors are convergent to zeros with H∞ performance index. A numerical simulation demonstrates the effectiveness of the proposed observers.
文摘The KT-II layer in the Zananor Oilfield,Caspian Basin,Kazakhstan,contains carbonate reservoirs of various types.The complex pore structure of the reservoirs have made it difficult to identify watered-out zones with traditional logging interpretation methods.This study classifies the reservoirs on the basis of core analysis and establishes an identification model for watered-out layers in the field to effectively improve the interpretation accuracy.Thin section analysis shows that there are three types of pores in the reservoirs,i.e.,the matrix pore,fracture and dissolution vug.A triple porosity model is used to calculate the porosities of the reservoirs and the results are combined with core analysis to classify the reservoirs into the fractured,matrix pore,fracture-pore as well as composite types.A classification standard is also proposed.There are differences in resistivity logging responses from the reservoirs of different types before and after watering-out.The preewatering-out resistivities are reconstructed using generalized neural network for different types of reservoirs.The watered-out layers can be effectively identified according to the difference in resistivity curves before and after watering-out.The results show that the watered-out layers identified with the method are consistent with measured data,thus serving as a reference for the evaluation of watered-out layers in the study area.
文摘In this paper, we improve traditional generative adversarial networks (GAN) with reference to residual networks and convolutional neural networks to facilitate the reconstruction of complex objects that cannot be reconstructed by traditional associative imaging methods. Unlike traditional ghost imaging to reconstruct objects from bucket signals, our proposed method can use simple objects (such as EMNIST) as a training set for GAN, and then recognize objects (such as faces) of completely different complexity than the training set. We use traditional ghost imaging and neural network to reconstruct target objects respectively. According to the research results in this paper, the method based on neural network can reconstruct complex objects very well, but the method based on traditional ghost imaging cannot reconstruct complex objects. The research scheme in this paper is of great significance for the reconstruction of complex object-related imaging under low sampling conditions.
基金Supported by 863 Program of China(2002AA2Z4291) Henan Innovation Project for University Prominent Research Talents(2005KYCX015)Henan Project for University Prominent Talents
文摘Proposed a new method to disclose the complicated non-linearity structure of the water-resource system, introducing chaos theory into the hydrology and water resources field, and combined with the chaos theory and artificial neural networks. Training data construction and networks structure were determined by the phase space reconstruction, and establishing nonlinear relationship of phase points with neural networks, the forecasting model of the resource quantity of the surface water was brought forward. The keystone of the way and the detailed arithmetic of the network training were given. The example shows that the model has highly forecasting precision.
文摘A new method of super-resolution image reconstruction is proposed, which uses a three-step-training error backpropagation neural network (BPNN) to realize the super-resolution reconstruction (SRR) of satellite image. The method is based on BPNN. First, three groups learning samples with different resolutions are obtained according to image observation model, and then vector mappings are respectively used to those three group learning samples to speed up the convergence of BPNN, at last, three times consecutive training are carried on the BPNN. Training samples used in each step are of higher resolution than those used in the previous steps, so the increasing weights store a great amount of information for SRR, and network performance and generalization ability are improved greatly. Simulation and generalization tests are carried on the well-trained three-step-training NN respectively, and the reconstruction results with higher resolution images verify the effectiveness and validity of this method.
文摘As digital image techniques have been widely used, the requirements for high-resolution images become increasingly stringent. Traditional single-frame interpolation techniques cannot add new high frequency information to the expanded images, and cannot improve resolution in deed. Multiframe-based techniques are effective ways for high-resolution image reconstruction, but their computation complexities and the difficulties in achieving image sequences limit their applications. An original method using an artificial neural network is proposed in this paper. Using the inherent merits in neural network, we can establish the mapping between high frequency components in low-resolution images and high-resolution images. Example applications and their results demonstrated the images reconstructed by our method are aesthetically and quantitatively (using the criteria of MSE and MAE) superior to the images acquired by common methods. Even for infrared images this method can give satisfactory results with high definition. In addition, a single-layer linear neural network is used in this paper, the computational complexity is very low, and this method can be realized in real time.
基金the National Natural Science Foundation of China(grant no.:81901958)Zhejiang Provincial National Natural Science Foundation of China(grant nos.:LY18H150004,LY19H150004,and LY20H150010).
文摘Background:Breast reconstruction is an effective technique to rebuild the appearance of the breasts in patients after mastectomy and improves the prognosis.The current study aimed to compare and analyze willingness for breast reconstruction after breast cancer between populations in China and the United States,from the perspective of social concern,using big data analysis.We also aimed to explore factors affecting surgical selection and to identify methods that can improve social cognition and acceptance of breast reconstruction.Methods:Using Baidu and Google,two representative Internet search engines in China and the United States as research tools,and using big data search volume as the benchmark,we compared and analyzed breast reconstruction willingness and attention characteristics between Chinese and American people,based on search heat,geographical distribution,age and sex,keyword distribution,ethnic group,and social development degree.Results:In both the long-term and short-term,Chinese people paid more attention towards searching about breast cancer,but less attention to breast reconstruction after breast cancer surgery.However,in both the short-term and long-term,people from the United States paid more attention towards breast cancer and breast reconstruction with the help of the Internet,showing a synchronous change relationship.There was a large regional difference in the search volume for breast cancer among the Chinese population,while no significant regional differences were noted in the search volume for breast cancer in the United States.However,a large regional difference was observed in the search volume for breast reconstruction between the two countries;people in the coastal and economically developed areas paid more attention to it.Most people who paid attention to breast reconstruction in China were women aged 20–39 years,while the attention among men was low.Search keywords were also limited to breast cancer-related information.However,between Asians and European Americans,Americans paid more attention to breast cancer and were affected by regional development,religious beliefs,and health facilities.Conclusion:Attention towards breast reconstruction after breast cancer was lower in the Chinese population than in the American population,and this difference was closely related to the level of regional development.There is insufficient information on breast reconstruction after breast cancer in recent Internet media.In addition to strengthening communication in clinics,media education is important to improve the cognitive level and social awareness of patients and their families,which is conducive to breast reconstruction.
文摘In robot-assisted surgery projects,researchers should be able to make fast 3D reconstruction. Usually 2D images acquired with common diagnostic equipments such as UT, CT and MRI are not enough and complete for an accurate 3D reconstruction. There are some interpolation methods for approximating non value voxels which consume large execution time. A novel algorithm is introduced based on generalized regression neural network (GRNN) which can interpolate unknown voxles fast and reliable. The GRNN interpolation is used to produce new 2D images between each two succeeding ultrasonic images. It is shown that the composition of GRNN with image distance transformation can produce higher quality 3D shapes. The results of this method are compared with other interpolation methods practically. It shows this method can decrease overall time consumption on online 3D reconstruction.
文摘Super-resolution reconstruction in medical imaging has become more demanding due to the necessity of obtaining high-quality images with minimal radiation dose,such as in low-field magnetic resonance imaging(MRI).However,image super-resolution reconstruction remains a difficult task because of the complexity and high textual requirements for diagnosis purpose.In this paper,we offer a deep learning based strategy for reconstructing medical images from low resolutions utilizing Transformer and generative adversarial networks(T-GANs).The integrated system can extract more precise texture information and focus more on important locations through global image matching after successfully inserting Transformer into the generative adversarial network for picture reconstruction.Furthermore,we weighted the combination of content loss,adversarial loss,and adversarial feature loss as the final multi-task loss function during the training of our proposed model T-GAN.In comparison to established measures like peak signal-to-noise ratio(PSNR)and structural similarity index measure(SSIM),our suggested T-GAN achieves optimal performance and recovers more texture features in super-resolution reconstruction of MRI scanned images of the knees and belly.