The degradation process of lithium-ion batteries is intricately linked to their entire lifecycle as power sources and energy storage devices,encompassing aspects such as performance delivery and cycling utilization.Co...The degradation process of lithium-ion batteries is intricately linked to their entire lifecycle as power sources and energy storage devices,encompassing aspects such as performance delivery and cycling utilization.Consequently,the accurate and expedient estimation or prediction of the aging state of lithium-ion batteries has garnered extensive attention.Nonetheless,prevailing research predominantly concentrates on either aging estimation or prediction,neglecting the dynamic fusion of both facets.This paper proposes a hybrid model for capacity aging estimation and prediction based on deep learning,wherein salient features highly pertinent to aging are extracted from charge and discharge relaxation processes.By amalgamating historical capacity decay data,the model dynamically furnishes estimations of the present capacity and forecasts of future capacity for lithium-ion batteries.Our approach is validated against a novel dataset involving charge and discharge cycles at varying rates.Specifically,under a charging condition of 0.25 C,a mean absolute percentage error(MAPE)of 0.29%is achieved.This outcome underscores the model's adeptness in harnessing relaxation processes commonly encountered in the real world and synergizing with historical capacity records within battery management systems(BMS),thereby affording estimations and prognostications of capacity decline with heightened precision.展开更多
Objective To observe the efficacy of deep learning(DL)model based on PET/CT and its combination with Cox proportional hazard model for predicting progressive disease(PD)of lung invasive adenocarcinoma within 5 years a...Objective To observe the efficacy of deep learning(DL)model based on PET/CT and its combination with Cox proportional hazard model for predicting progressive disease(PD)of lung invasive adenocarcinoma within 5 years after surgery.Methods The clinical,PET/CT and 5-year follow-up data of 250 patients with lung invasive adenocarcinoma were retrospectively analyzed.According to PD or not,the patients were divided into the PD group(n=71)and non-PD group(n=179).The basic data and PET/CT findings were compared between groups,among which the quantitative variables being significant different between groups were transformed to categorical variables using receiver operating characteristic(ROC)curve and corresponding cut-off value.Multivariant Cox proportional hazard model was used to select independent predicting factors of PD of lung invasive adenocarcinoma within 5 years after surgery.The patients were divided into training,validation and test sets at the ratio of 6∶2∶2,and PET/CT data in training set and validation set were used to train model and tuning parameters to build the PET/CT DL model,and the combination model was built in serial connection of DL model and the predictive factors.In test set,the efficacy of each model for predicting PD of lung invasive adenocarcinoma within 5 years after surgery was assessed and compared using the area under the curve(AUC).Results Patients'gender and smoking status,as well as the long diameter,SUV max and SUV mean of lesions measured on PET images,the long diameter,short diameter and type of lesions showed on CT were statistically different between groups(all P<0.05).Smoking(HR=1.787[1.053,3.031],P=0.031)and lesion SUV max>4.15(HR=5.249[1.062,25.945],P=0.042)were both predictors of PD of lung invasive adenocarcinoma within 5 years after surgery.In test set,the AUC of PET/CT DL model for predicting PD was 0.847,of the combination model was 0.890,of the latter was higher than of the former(P=0.036).Conclusion DL model based on PET/CT had high efficacy for predicting PD of lung invasive adenocarcinoma within 5 years after surgery.Combining with Cox proportional hazard model could further improve its predicting efficacy.展开更多
Landfill leaks pose a serious threat to environmental health,risking the contamination of both groundwater and soil resources.Accurate investigation of these sites is essential for implementing effective prevention an...Landfill leaks pose a serious threat to environmental health,risking the contamination of both groundwater and soil resources.Accurate investigation of these sites is essential for implementing effective prevention and control measures.The self-potential(SP)stands out for its sensitivity to contamination plumes,offering a solution for monitoring and detecting the movement and seepage of subsurface pollutants.However,traditional SP inversion techniques heavily rely on precise subsurface resistivity information.In this study,we propose the Attention U-Net deep learning network for rapid SP inversion.By incorporating an attention mechanism,this algorithm effectively learns the relationship between array-style SP data and the location and extent of subsurface contaminated sources.We designed a synthetic landfill model with a heterogeneous resistivity structure to assess the performance of Attention U-Net deep learning network.Additionally,we conducted further validation using a laboratory model to assess its practical applicability.The results demonstrate that the algorithm is not solely dependent on resistivity information,enabling effective locating of the source distribution,even in models with intricate subsurface structures.Our work provides a promising tool for SP data processing,enhancing the applicability of this method in the field of near-subsurface environmental monitoring.展开更多
Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices...Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices,and it is not environmental-friendly with much power cost.In this paper,we focus on low-rank optimization for efficient deep learning techniques.In the space domain,DNNs are compressed by low rank approximation of the network parameters,which directly reduces the storage requirement with a smaller number of network parameters.In the time domain,the network parameters can be trained in a few subspaces,which enables efficient training for fast convergence.The model compression in the spatial domain is summarized into three categories as pre-train,pre-set,and compression-aware methods,respectively.With a series of integrable techniques discussed,such as sparse pruning,quantization,and entropy coding,we can ensemble them in an integration framework with lower computational complexity and storage.In addition to summary of recent technical advances,we have two findings for motivating future works.One is that the effective rank,derived from the Shannon entropy of the normalized singular values,outperforms other conventional sparse measures such as the?_1 norm for network compression.The other is a spatial and temporal balance for tensorized neural networks.For accelerating the training of tensorized neural networks,it is crucial to leverage redundancy for both model compression and subspace training.展开更多
Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-cham...Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-chamber views two-dimensional echocardiograms were obtained prospectively in 205 patients with coronary heart disease.The model for evaluating LV regional contractile function was constructed using a five-fold cross-validation method to automatically identify the presence of RWMA or not,and the performance of this model was assessed taken manual interpretation of RWMA as standards.Results Among 205 patients,RWMA was detected in totally 650 segments in 83 cases.LV myocardial segmentation model demonstrated good efficacy for delineation of LV myocardium.The average Dice similarity coefficient for LV myocardial segmentation results in the apical two-chamber,three-chamber and four-chamber views was 0.85,0.82 and 0.88,respectively.LV myocardial segmentation model accurately segmented LV myocardium in apical two-chamber,three-chamber and four-chamber views.The mean area under the curve(AUC)of RWMA identification model was 0.843±0.071,with sensitivity of(64.19±14.85)%,specificity of(89.44±7.31)%and accuracy of(85.22±4.37)%.Conclusion Deep learning echocardiographic intelligent model could be used to automatically evaluate LV regional contractile function,hence rapidly and accurately identifying RWMA.展开更多
Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,...Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.展开更多
[Objective]Urban floods are occurring more frequently because of global climate change and urbanization.Accordingly,urban rainstorm and flood forecasting has become a priority in urban hydrology research.However,two-d...[Objective]Urban floods are occurring more frequently because of global climate change and urbanization.Accordingly,urban rainstorm and flood forecasting has become a priority in urban hydrology research.However,two-dimensional hydrodynamic models execute calculations slowly,hindering the rapid simulation and forecasting of urban floods.To overcome this limitation and accelerate the speed and improve the accuracy of urban flood simulations and forecasting,numerical simulations and deep learning were combined to develop a more effective urban flood forecasting method.[Methods]Specifically,a cellular automata model was used to simulate the urban flood process and address the need to include a large number of datasets in the deep learning process.Meanwhile,to shorten the time required for urban flood forecasting,a convolutional neural network model was used to establish the mapping relationship between rainfall and inundation depth.[Results]The results show that the relative error of forecasting the maximum inundation depth in flood-prone locations is less than 10%,and the Nash efficiency coefficient of forecasting inundation depth series in flood-prone locations is greater than 0.75.[Conclusion]The result demonstrated that the proposed method could execute highly accurate simulations and quickly produce forecasts,illustrating its superiority as an urban flood forecasting technique.展开更多
Deep learning has achieved excellent results in various tasks in the field of computer vision,especially in fine-grained visual categorization.It aims to distinguish the subordinate categories of the label-level categ...Deep learning has achieved excellent results in various tasks in the field of computer vision,especially in fine-grained visual categorization.It aims to distinguish the subordinate categories of the label-level categories.Due to high intra-class variances and high inter-class similarity,the fine-grained visual categorization is extremely challenging.This paper first briefly introduces and analyzes the related public datasets.After that,some of the latest methods are reviewed.Based on the feature types,the feature processing methods,and the overall structure used in the model,we divide them into three types of methods:methods based on general convolutional neural network(CNN)and strong supervision of parts,methods based on single feature processing,and meth-ods based on multiple feature processing.Most methods of the first type have a relatively simple structure,which is the result of the initial research.The methods of the other two types include models that have special structures and training processes,which are helpful to obtain discriminative features.We conduct a specific analysis on several methods with high accuracy on pub-lic datasets.In addition,we support that the focus of the future research is to solve the demand of existing methods for the large amount of the data and the computing power.In terms of tech-nology,the extraction of the subtle feature information with the burgeoning vision transformer(ViT)network is also an important research direction.展开更多
According to the oversampling imaging characteristics, an infrared small target detection method based on deep learning is proposed. A 7-layer deep convolutional neural network(CNN) is designed to automatically extrac...According to the oversampling imaging characteristics, an infrared small target detection method based on deep learning is proposed. A 7-layer deep convolutional neural network(CNN) is designed to automatically extract small target features and suppress clutters in an end-to-end manner. The input of CNN is an original oversampling image while the output is a cluttersuppressed feature map. The CNN contains only convolution and non-linear operations, and the resolution of the output feature map is the same as that of the input image. The L1-norm loss function is used, and a mass of training data is generated to train the network effectively. Results show that compared with several baseline methods, the proposed method improves the signal clutter ratio gain and background suppression factor by 3–4 orders of magnitude, and has more powerful target detection performance.展开更多
The accurate and efficient prediction of explosive detonation properties has important engineering significance for weapon design.Traditional methods for predicting detonation performance include empirical formulas,eq...The accurate and efficient prediction of explosive detonation properties has important engineering significance for weapon design.Traditional methods for predicting detonation performance include empirical formulas,equations of state,and quantum chemical calculation methods.In recent years,with the development of computer performance and deep learning methods,researchers have begun to apply deep learning methods to the prediction of explosive detonation performance.The deep learning method has the advantage of simple and rapid prediction of explosive detonation properties.However,some problems remain in the study of detonation properties based on deep learning.For example,there are few studies on the prediction of mixed explosives,on the prediction of the parameters of the equation of state of explosives,and on the application of explosive properties to predict the formulation of explosives.Based on an artificial neural network model and a one-dimensional convolutional neural network model,three improved deep learning models were established in this work with the aim of solving these problems.The training data for these models,called the detonation parameters prediction model,JWL equation of state(EOS)prediction model,and inverse prediction model,was obtained through the KHT thermochemical code.After training,the model was tested for overfitting using the validation-set test.Through the model-accuracy test,the prediction accuracy of the model for real explosive formulations was tested by comparing the predicted value with the reference value.The results show that the model errors were within 10%and 3%for the prediction of detonation pressure and detonation velocity,respectively.The accuracy refers to the prediction of tested explosive formulations which consist of TNT,RDX and HMX.For the prediction of the equation of state for explosives,the correlation coefficient between the prediction and the reference curves was above 0.99.For the prediction of the inverse prediction model,the prediction error of the explosive equation was within 9%.This indicates that the models have utility in engineering.展开更多
It is generally believed that intelligent management for sewage treatment plants(STPs) is essential to the sustainable engineering of future smart cities.The core of management lies in the precise prediction of daily ...It is generally believed that intelligent management for sewage treatment plants(STPs) is essential to the sustainable engineering of future smart cities.The core of management lies in the precise prediction of daily volumes of sewage.The generation of sewage is the result of multiple factors from the whole social system.Characterized by strong process abstraction ability,data mining techniques have been viewed as promising prediction methods to realize intelligent STP management.However,existing data mining-based methods for this purpose just focus on a single factor such as an economical or meteorological factor and ignore their collaborative effects.To address this challenge,a deep learning-based intelligent management mechanism for STPs is proposed,to predict business volume.Specifically,the grey relation algorithm(GRA) and gated recursive unit network(GRU) are combined into a prediction model(GRAGRU).The GRA is utilized to select the factors that have a significant impact on the sewage business volume,and the GRU is set up to output the prediction results.We conducted a large number of experiments to verify the efficiency of the proposed GRA-GRU model.展开更多
The issue of small-angle maneuvering targets inverse synthetic aperture radar(ISAR)imaging has been successfully addressed by popular motion compensation algorithms.However,when the target’s rotational velocity is su...The issue of small-angle maneuvering targets inverse synthetic aperture radar(ISAR)imaging has been successfully addressed by popular motion compensation algorithms.However,when the target’s rotational velocity is sufficiently high during the dwell time of the radar,such compensation algorithms cannot obtain a high quality image.This paper proposes an ISAR imaging algorithm based on keystone transform and deep learning algorithm.The keystone transform is used to coarsely compensate for the target’s rotational motion and translational motion,and the deep learning algorithm is used to achieve a super-resolution image.The uniformly distributed point target data are used as the data set of the training u-net network.In addition,this method does not require estimating the motion parameters of the target,which simplifies the algorithm steps.Finally,several experiments are performed to demonstrate the effectiveness of the proposed algorithm.展开更多
Channel estimation has been considered as a key issue in the millimeter-wave(mmWave)massive multi-input multioutput(MIMO)communication systems,which becomes more challenging with a large number of antennas.In this pap...Channel estimation has been considered as a key issue in the millimeter-wave(mmWave)massive multi-input multioutput(MIMO)communication systems,which becomes more challenging with a large number of antennas.In this paper,we propose a deep learning(DL)-based fast channel estimation method for mmWave massive MIMO systems.The proposed method can directly and effectively estimate channel state information(CSI)from received data without performing pilot signals estimate in advance,which simplifies the estimation process.Specifically,we develop a convolutional neural network(CNN)-based channel estimation network for the case of dimensional mismatch of input and output data,subsequently denoted as channel(H)neural network(HNN).It can quickly estimate the channel information by learning the inherent characteristics of the received data and the relationship between the received data and the channel,while the dimension of the received data is much smaller than the channel matrix.Simulation results show that the proposed HNN can gain better channel estimation accuracy compared with existing schemes.展开更多
Nyquist Folding Receiver(NYFR)is a perceptron structure that realizes a low probability of intercept(LPI)signal analog to information.Aiming at the problem of LPI radar signal receiving,the time domain,frequency domai...Nyquist Folding Receiver(NYFR)is a perceptron structure that realizes a low probability of intercept(LPI)signal analog to information.Aiming at the problem of LPI radar signal receiving,the time domain,frequency domain,and time-frequency domain problems of signals intercepted by NYFR structure are studied.Combined with the time-frequency analysis(TFA)method,a radar recognition scheme based on deep learning(DL)is introduced,which can reliably classify common LPI radar signals.First,the structure of NYFR and its characteristics in the time domain,frequency domain,and time and frequency domain are analyzed.Then,the received signal is then converted into a time-frequency image(TFI).Finally,four kinds of DL algorithms are used to classify LPI radar signals.Simulation results demonstrate the correctness of the NYFR structure,and the effectiveness of the proposed recognition method is verified by comparison experiments.展开更多
As the representative of flexibility in optical imaging media,in recent years,fiber bundles have emerged as a promising architecture in the development of compact visual systems.Dedicated to tackling the problems of u...As the representative of flexibility in optical imaging media,in recent years,fiber bundles have emerged as a promising architecture in the development of compact visual systems.Dedicated to tackling the problems of universal honeycomb artifacts and low signal-to-noise ratio(SNR)imaging in fiber bundles,the iterative super-resolution reconstruction network based on a physical model is proposed.Under the constraint of solving the two subproblems of data fidelity and prior regularization term alternately,the network can efficiently“regenerate”the lost spatial resolution with deep learning.By building and calibrating a dual-path imaging system,the real-world dataset where paired low-resolution(LR)-high-resolution(HR)images on the same scene can be generated simultaneously.Numerical results on both the United States Air Force(USAF)resolution target and complex target objects demonstrate that the algorithm can restore high-contrast images without pixilated noise.On the basis of super-resolution reconstruction,compound eye image composition based on fiber bundle is also embedded in this paper for the actual imaging requirements.The proposed work is the first to apply a physical model-based deep learning network to fiber bundle imaging in the infrared band,effectively promoting the engineering application of thermal radiation detection.展开更多
Social infrastructures such as dams are likely to be exposed to high risk of terrorist and military attacks,leading to increasing attentions on their vulnerability and catastrophic consequences under such events.This ...Social infrastructures such as dams are likely to be exposed to high risk of terrorist and military attacks,leading to increasing attentions on their vulnerability and catastrophic consequences under such events.This paper tries to develop advanced deep learning approaches for structural dynamic response prediction and dam health diagnosis.At first,the improved long short-term memory(LSTM)networks are proposed for data-driven structural dynamic response analysis with the data generated by a single degree of freedom(SDOF)and the finite numerical simulation,due to the unavailability of abundant practical structural response data of concrete gravity dam under blast events.Three kinds of LSTM-based models are discussed with the various cases of noise-contaminated signals,and the results prove that LSTM-based models have the potential for quick structural response estimation under blast loads.Furthermore,the damage indicators(i.e.,peak vibration velocity and domain frequency)are extracted from the predicted velocity histories,and their relationship with the dam damage status from the numerical simulation is established.This study provides a deep-learning based structural health monitoring(SHM)framework for quick assessment of dam experienced underwater explosions through blastinduced monitoring data.展开更多
With the warming up and continuous development of machine learning,especially deep learning,the research on visual question answering field has made significant progress,with important theoretical research significanc...With the warming up and continuous development of machine learning,especially deep learning,the research on visual question answering field has made significant progress,with important theoretical research significance and practical application value.Therefore,it is necessary to summarize the current research and provide some reference for researchers in this field.This article conducted a detailed and in-depth analysis and summarized of relevant research and typical methods of visual question answering field.First,relevant background knowledge about VQA(Visual Question Answering)was introduced.Secondly,the issues and challenges of visual question answering were discussed,and at the same time,some promising discussion on the particular methodologies was given.Thirdly,the key sub-problems affecting visual question answering were summarized and analyzed.Then,the current commonly used data sets and evaluation indicators were summarized.Next,in view of the popular algorithms and models in VQA research,comparison of the algorithms and models was summarized and listed.Finally,the future development trend and conclusion of visual question answering were prospected.展开更多
An obstacle perception system for intelligent vehicle is proposed.The proposed system combines the stereo version technique and the deep learning network model,and is applied to obstacle perception tasks in complex en...An obstacle perception system for intelligent vehicle is proposed.The proposed system combines the stereo version technique and the deep learning network model,and is applied to obstacle perception tasks in complex environment.In this paper,we provide a complete system design project,which includes the hardware parameters,software framework,algorithm principle,and optimization method.In addition,special experiments are designed to demonstrate that the performance of the proposed system meets the requirements of actual application.The experiment results show that the proposed system is valid to both standard obstacles and non-standard obstacles,and suitable for different weather and lighting conditions in complex environment.It announces that the proposed system is flexible and robust to the intelligent vehicle.展开更多
To address the issue of neglecting scenarios involving joint operations and collaborative drone swarm operations in air combat target intent recognition.This paper proposes a transfer learning-based intention predicti...To address the issue of neglecting scenarios involving joint operations and collaborative drone swarm operations in air combat target intent recognition.This paper proposes a transfer learning-based intention prediction model for drone formation targets in air combat.This model recognizes the intentions of multiple aerial targets by extracting spatial features among the targets at each moment.Simulation results demonstrate that,compared to classical intention recognition models,the proposed model in this paper achieves higher accuracy in identifying the intentions of drone swarm targets in air combat scenarios.展开更多
Autonomous umanned aerial vehicle(UAV) manipulation is necessary for the defense department to execute tactical missions given by commanders in the future unmanned battlefield. A large amount of research has been devo...Autonomous umanned aerial vehicle(UAV) manipulation is necessary for the defense department to execute tactical missions given by commanders in the future unmanned battlefield. A large amount of research has been devoted to improving the autonomous decision-making ability of UAV in an interactive environment, where finding the optimal maneuvering decisionmaking policy became one of the key issues for enabling the intelligence of UAV. In this paper, we propose a maneuvering decision-making algorithm for autonomous air-delivery based on deep reinforcement learning under the guidance of expert experience. Specifically, we refine the guidance towards area and guidance towards specific point tasks for the air-delivery process based on the traditional air-to-surface fire control methods.Moreover, we construct the UAV maneuvering decision-making model based on Markov decision processes(MDPs). Specifically, we present a reward shaping method for the guidance towards area and guidance towards specific point tasks using potential-based function and expert-guided advice. The proposed algorithm could accelerate the convergence of the maneuvering decision-making policy and increase the stability of the policy in terms of the output during the later stage of training process. The effectiveness of the proposed maneuvering decision-making policy is illustrated by the curves of training parameters and extensive experimental results for testing the trained policy.展开更多
文摘The degradation process of lithium-ion batteries is intricately linked to their entire lifecycle as power sources and energy storage devices,encompassing aspects such as performance delivery and cycling utilization.Consequently,the accurate and expedient estimation or prediction of the aging state of lithium-ion batteries has garnered extensive attention.Nonetheless,prevailing research predominantly concentrates on either aging estimation or prediction,neglecting the dynamic fusion of both facets.This paper proposes a hybrid model for capacity aging estimation and prediction based on deep learning,wherein salient features highly pertinent to aging are extracted from charge and discharge relaxation processes.By amalgamating historical capacity decay data,the model dynamically furnishes estimations of the present capacity and forecasts of future capacity for lithium-ion batteries.Our approach is validated against a novel dataset involving charge and discharge cycles at varying rates.Specifically,under a charging condition of 0.25 C,a mean absolute percentage error(MAPE)of 0.29%is achieved.This outcome underscores the model's adeptness in harnessing relaxation processes commonly encountered in the real world and synergizing with historical capacity records within battery management systems(BMS),thereby affording estimations and prognostications of capacity decline with heightened precision.
文摘Objective To observe the efficacy of deep learning(DL)model based on PET/CT and its combination with Cox proportional hazard model for predicting progressive disease(PD)of lung invasive adenocarcinoma within 5 years after surgery.Methods The clinical,PET/CT and 5-year follow-up data of 250 patients with lung invasive adenocarcinoma were retrospectively analyzed.According to PD or not,the patients were divided into the PD group(n=71)and non-PD group(n=179).The basic data and PET/CT findings were compared between groups,among which the quantitative variables being significant different between groups were transformed to categorical variables using receiver operating characteristic(ROC)curve and corresponding cut-off value.Multivariant Cox proportional hazard model was used to select independent predicting factors of PD of lung invasive adenocarcinoma within 5 years after surgery.The patients were divided into training,validation and test sets at the ratio of 6∶2∶2,and PET/CT data in training set and validation set were used to train model and tuning parameters to build the PET/CT DL model,and the combination model was built in serial connection of DL model and the predictive factors.In test set,the efficacy of each model for predicting PD of lung invasive adenocarcinoma within 5 years after surgery was assessed and compared using the area under the curve(AUC).Results Patients'gender and smoking status,as well as the long diameter,SUV max and SUV mean of lesions measured on PET images,the long diameter,short diameter and type of lesions showed on CT were statistically different between groups(all P<0.05).Smoking(HR=1.787[1.053,3.031],P=0.031)and lesion SUV max>4.15(HR=5.249[1.062,25.945],P=0.042)were both predictors of PD of lung invasive adenocarcinoma within 5 years after surgery.In test set,the AUC of PET/CT DL model for predicting PD was 0.847,of the combination model was 0.890,of the latter was higher than of the former(P=0.036).Conclusion DL model based on PET/CT had high efficacy for predicting PD of lung invasive adenocarcinoma within 5 years after surgery.Combining with Cox proportional hazard model could further improve its predicting efficacy.
基金Projects(42174170,41874145,72088101)supported by the National Natural Science Foundation of ChinaProject(CX20200228)supported by the Hunan Provincial Innovation Foundation for Postgraduate,China。
文摘Landfill leaks pose a serious threat to environmental health,risking the contamination of both groundwater and soil resources.Accurate investigation of these sites is essential for implementing effective prevention and control measures.The self-potential(SP)stands out for its sensitivity to contamination plumes,offering a solution for monitoring and detecting the movement and seepage of subsurface pollutants.However,traditional SP inversion techniques heavily rely on precise subsurface resistivity information.In this study,we propose the Attention U-Net deep learning network for rapid SP inversion.By incorporating an attention mechanism,this algorithm effectively learns the relationship between array-style SP data and the location and extent of subsurface contaminated sources.We designed a synthetic landfill model with a heterogeneous resistivity structure to assess the performance of Attention U-Net deep learning network.Additionally,we conducted further validation using a laboratory model to assess its practical applicability.The results demonstrate that the algorithm is not solely dependent on resistivity information,enabling effective locating of the source distribution,even in models with intricate subsurface structures.Our work provides a promising tool for SP data processing,enhancing the applicability of this method in the field of near-subsurface environmental monitoring.
基金supported by the National Natural Science Foundation of China(62171088,U19A2052,62020106011)the Medico-Engineering Cooperation Funds from University of Electronic Science and Technology of China(ZYGX2021YGLH215,ZYGX2022YGRH005)。
文摘Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices,and it is not environmental-friendly with much power cost.In this paper,we focus on low-rank optimization for efficient deep learning techniques.In the space domain,DNNs are compressed by low rank approximation of the network parameters,which directly reduces the storage requirement with a smaller number of network parameters.In the time domain,the network parameters can be trained in a few subspaces,which enables efficient training for fast convergence.The model compression in the spatial domain is summarized into three categories as pre-train,pre-set,and compression-aware methods,respectively.With a series of integrable techniques discussed,such as sparse pruning,quantization,and entropy coding,we can ensemble them in an integration framework with lower computational complexity and storage.In addition to summary of recent technical advances,we have two findings for motivating future works.One is that the effective rank,derived from the Shannon entropy of the normalized singular values,outperforms other conventional sparse measures such as the?_1 norm for network compression.The other is a spatial and temporal balance for tensorized neural networks.For accelerating the training of tensorized neural networks,it is crucial to leverage redundancy for both model compression and subspace training.
文摘Objective To observe the value of deep learning echocardiographic intelligent model for evaluation on left ventricular(LV)regional wall motion abnormalities(RWMA).Methods Apical two-chamber,three-chamber and four-chamber views two-dimensional echocardiograms were obtained prospectively in 205 patients with coronary heart disease.The model for evaluating LV regional contractile function was constructed using a five-fold cross-validation method to automatically identify the presence of RWMA or not,and the performance of this model was assessed taken manual interpretation of RWMA as standards.Results Among 205 patients,RWMA was detected in totally 650 segments in 83 cases.LV myocardial segmentation model demonstrated good efficacy for delineation of LV myocardium.The average Dice similarity coefficient for LV myocardial segmentation results in the apical two-chamber,three-chamber and four-chamber views was 0.85,0.82 and 0.88,respectively.LV myocardial segmentation model accurately segmented LV myocardium in apical two-chamber,three-chamber and four-chamber views.The mean area under the curve(AUC)of RWMA identification model was 0.843±0.071,with sensitivity of(64.19±14.85)%,specificity of(89.44±7.31)%and accuracy of(85.22±4.37)%.Conclusion Deep learning echocardiographic intelligent model could be used to automatically evaluate LV regional contractile function,hence rapidly and accurately identifying RWMA.
基金supported by the National Natural Science Foundation of China(61877067)the Foundation of Science and Technology on Near-Surface Detection Laboratory(TCGZ2019A002,TCGZ2021C003,6142414200511)the Natural Science Basic Research Program of Shaanxi(2021JZ-19)。
文摘Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.
文摘[Objective]Urban floods are occurring more frequently because of global climate change and urbanization.Accordingly,urban rainstorm and flood forecasting has become a priority in urban hydrology research.However,two-dimensional hydrodynamic models execute calculations slowly,hindering the rapid simulation and forecasting of urban floods.To overcome this limitation and accelerate the speed and improve the accuracy of urban flood simulations and forecasting,numerical simulations and deep learning were combined to develop a more effective urban flood forecasting method.[Methods]Specifically,a cellular automata model was used to simulate the urban flood process and address the need to include a large number of datasets in the deep learning process.Meanwhile,to shorten the time required for urban flood forecasting,a convolutional neural network model was used to establish the mapping relationship between rainfall and inundation depth.[Results]The results show that the relative error of forecasting the maximum inundation depth in flood-prone locations is less than 10%,and the Nash efficiency coefficient of forecasting inundation depth series in flood-prone locations is greater than 0.75.[Conclusion]The result demonstrated that the proposed method could execute highly accurate simulations and quickly produce forecasts,illustrating its superiority as an urban flood forecasting technique.
基金supported by the National Natural Science Foundation of China(61571453,61806218).
文摘Deep learning has achieved excellent results in various tasks in the field of computer vision,especially in fine-grained visual categorization.It aims to distinguish the subordinate categories of the label-level categories.Due to high intra-class variances and high inter-class similarity,the fine-grained visual categorization is extremely challenging.This paper first briefly introduces and analyzes the related public datasets.After that,some of the latest methods are reviewed.Based on the feature types,the feature processing methods,and the overall structure used in the model,we divide them into three types of methods:methods based on general convolutional neural network(CNN)and strong supervision of parts,methods based on single feature processing,and meth-ods based on multiple feature processing.Most methods of the first type have a relatively simple structure,which is the result of the initial research.The methods of the other two types include models that have special structures and training processes,which are helpful to obtain discriminative features.We conduct a specific analysis on several methods with high accuracy on pub-lic datasets.In addition,we support that the focus of the future research is to solve the demand of existing methods for the large amount of the data and the computing power.In terms of tech-nology,the extraction of the subtle feature information with the burgeoning vision transformer(ViT)network is also an important research direction.
基金supported by the National Key Research and Development Program of China(2016YFB0500901)the Natural Science Foundation of Shanghai(18ZR1437200)the Satellite Mapping Technology and Application National Key Laboratory of Geographical Information Bureau(KLSMTA-201709)
文摘According to the oversampling imaging characteristics, an infrared small target detection method based on deep learning is proposed. A 7-layer deep convolutional neural network(CNN) is designed to automatically extract small target features and suppress clutters in an end-to-end manner. The input of CNN is an original oversampling image while the output is a cluttersuppressed feature map. The CNN contains only convolution and non-linear operations, and the resolution of the output feature map is the same as that of the input image. The L1-norm loss function is used, and a mass of training data is generated to train the network effectively. Results show that compared with several baseline methods, the proposed method improves the signal clutter ratio gain and background suppression factor by 3–4 orders of magnitude, and has more powerful target detection performance.
文摘The accurate and efficient prediction of explosive detonation properties has important engineering significance for weapon design.Traditional methods for predicting detonation performance include empirical formulas,equations of state,and quantum chemical calculation methods.In recent years,with the development of computer performance and deep learning methods,researchers have begun to apply deep learning methods to the prediction of explosive detonation performance.The deep learning method has the advantage of simple and rapid prediction of explosive detonation properties.However,some problems remain in the study of detonation properties based on deep learning.For example,there are few studies on the prediction of mixed explosives,on the prediction of the parameters of the equation of state of explosives,and on the application of explosive properties to predict the formulation of explosives.Based on an artificial neural network model and a one-dimensional convolutional neural network model,three improved deep learning models were established in this work with the aim of solving these problems.The training data for these models,called the detonation parameters prediction model,JWL equation of state(EOS)prediction model,and inverse prediction model,was obtained through the KHT thermochemical code.After training,the model was tested for overfitting using the validation-set test.Through the model-accuracy test,the prediction accuracy of the model for real explosive formulations was tested by comparing the predicted value with the reference value.The results show that the model errors were within 10%and 3%for the prediction of detonation pressure and detonation velocity,respectively.The accuracy refers to the prediction of tested explosive formulations which consist of TNT,RDX and HMX.For the prediction of the equation of state for explosives,the correlation coefficient between the prediction and the reference curves was above 0.99.For the prediction of the inverse prediction model,the prediction error of the explosive equation was within 9%.This indicates that the models have utility in engineering.
基金Project(KJZD-M202000801) supported by the Major Project of Chongqing Municipal Education Commission,ChinaProject(2016YFE0205600) supported by the National Key Research&Development Program of China+1 种基金Project(CXQT19023) supported by the Chongqing University Innovation Group Project,ChinaProjects(KFJJ2018069,1853061,1856033) supported by the Key Platform Opening Project of Chongqing Technology and Business University,China。
文摘It is generally believed that intelligent management for sewage treatment plants(STPs) is essential to the sustainable engineering of future smart cities.The core of management lies in the precise prediction of daily volumes of sewage.The generation of sewage is the result of multiple factors from the whole social system.Characterized by strong process abstraction ability,data mining techniques have been viewed as promising prediction methods to realize intelligent STP management.However,existing data mining-based methods for this purpose just focus on a single factor such as an economical or meteorological factor and ignore their collaborative effects.To address this challenge,a deep learning-based intelligent management mechanism for STPs is proposed,to predict business volume.Specifically,the grey relation algorithm(GRA) and gated recursive unit network(GRU) are combined into a prediction model(GRAGRU).The GRA is utilized to select the factors that have a significant impact on the sewage business volume,and the GRU is set up to output the prediction results.We conducted a large number of experiments to verify the efficiency of the proposed GRA-GRU model.
基金This work was supported by the National Natural Science Foundation of China(61571388,61871465,62071414)the Project of Introducing Overseas Students in Hebei Province(C20200367).
文摘The issue of small-angle maneuvering targets inverse synthetic aperture radar(ISAR)imaging has been successfully addressed by popular motion compensation algorithms.However,when the target’s rotational velocity is sufficiently high during the dwell time of the radar,such compensation algorithms cannot obtain a high quality image.This paper proposes an ISAR imaging algorithm based on keystone transform and deep learning algorithm.The keystone transform is used to coarsely compensate for the target’s rotational motion and translational motion,and the deep learning algorithm is used to achieve a super-resolution image.The uniformly distributed point target data are used as the data set of the training u-net network.In addition,this method does not require estimating the motion parameters of the target,which simplifies the algorithm steps.Finally,several experiments are performed to demonstrate the effectiveness of the proposed algorithm.
基金supported by the National Key R&D Program of China(2018YFB1802004)111 Project(B08038)。
文摘Channel estimation has been considered as a key issue in the millimeter-wave(mmWave)massive multi-input multioutput(MIMO)communication systems,which becomes more challenging with a large number of antennas.In this paper,we propose a deep learning(DL)-based fast channel estimation method for mmWave massive MIMO systems.The proposed method can directly and effectively estimate channel state information(CSI)from received data without performing pilot signals estimate in advance,which simplifies the estimation process.Specifically,we develop a convolutional neural network(CNN)-based channel estimation network for the case of dimensional mismatch of input and output data,subsequently denoted as channel(H)neural network(HNN).It can quickly estimate the channel information by learning the inherent characteristics of the received data and the relationship between the received data and the channel,while the dimension of the received data is much smaller than the channel matrix.Simulation results show that the proposed HNN can gain better channel estimation accuracy compared with existing schemes.
基金supported by the National Defence Pre-research Foundation of China。
文摘Nyquist Folding Receiver(NYFR)is a perceptron structure that realizes a low probability of intercept(LPI)signal analog to information.Aiming at the problem of LPI radar signal receiving,the time domain,frequency domain,and time-frequency domain problems of signals intercepted by NYFR structure are studied.Combined with the time-frequency analysis(TFA)method,a radar recognition scheme based on deep learning(DL)is introduced,which can reliably classify common LPI radar signals.First,the structure of NYFR and its characteristics in the time domain,frequency domain,and time and frequency domain are analyzed.Then,the received signal is then converted into a time-frequency image(TFI).Finally,four kinds of DL algorithms are used to classify LPI radar signals.Simulation results demonstrate the correctness of the NYFR structure,and the effectiveness of the proposed recognition method is verified by comparison experiments.
基金the National Natural Science Foundation of China(Grant Nos.61905115,62105151,62175109,U21B2033)Leading Technology of Jiangsu Basic Research Plan(Grant No.BK20192003)+2 种基金Youth Foundation of Jiangsu Province(Grant Nos.BK20190445,BK20210338)Fundamental Research Funds for the Central Universities(Grant No.30920032101)Open Research Fund of Jiangsu Key Laboratory of Spectral Imaging&Intelligent Sense(Grant No.JSGP202105)to provide fund for conducting experiments。
文摘As the representative of flexibility in optical imaging media,in recent years,fiber bundles have emerged as a promising architecture in the development of compact visual systems.Dedicated to tackling the problems of universal honeycomb artifacts and low signal-to-noise ratio(SNR)imaging in fiber bundles,the iterative super-resolution reconstruction network based on a physical model is proposed.Under the constraint of solving the two subproblems of data fidelity and prior regularization term alternately,the network can efficiently“regenerate”the lost spatial resolution with deep learning.By building and calibrating a dual-path imaging system,the real-world dataset where paired low-resolution(LR)-high-resolution(HR)images on the same scene can be generated simultaneously.Numerical results on both the United States Air Force(USAF)resolution target and complex target objects demonstrate that the algorithm can restore high-contrast images without pixilated noise.On the basis of super-resolution reconstruction,compound eye image composition based on fiber bundle is also embedded in this paper for the actual imaging requirements.The proposed work is the first to apply a physical model-based deep learning network to fiber bundle imaging in the infrared band,effectively promoting the engineering application of thermal radiation detection.
基金supported by a grant from the National Natural Science Foundation of China(Grant No.52109163 and 51979188).
文摘Social infrastructures such as dams are likely to be exposed to high risk of terrorist and military attacks,leading to increasing attentions on their vulnerability and catastrophic consequences under such events.This paper tries to develop advanced deep learning approaches for structural dynamic response prediction and dam health diagnosis.At first,the improved long short-term memory(LSTM)networks are proposed for data-driven structural dynamic response analysis with the data generated by a single degree of freedom(SDOF)and the finite numerical simulation,due to the unavailability of abundant practical structural response data of concrete gravity dam under blast events.Three kinds of LSTM-based models are discussed with the various cases of noise-contaminated signals,and the results prove that LSTM-based models have the potential for quick structural response estimation under blast loads.Furthermore,the damage indicators(i.e.,peak vibration velocity and domain frequency)are extracted from the predicted velocity histories,and their relationship with the dam damage status from the numerical simulation is established.This study provides a deep-learning based structural health monitoring(SHM)framework for quick assessment of dam experienced underwater explosions through blastinduced monitoring data.
基金Project(61702063)supported by the National Natural Science Foundation of China。
文摘With the warming up and continuous development of machine learning,especially deep learning,the research on visual question answering field has made significant progress,with important theoretical research significance and practical application value.Therefore,it is necessary to summarize the current research and provide some reference for researchers in this field.This article conducted a detailed and in-depth analysis and summarized of relevant research and typical methods of visual question answering field.First,relevant background knowledge about VQA(Visual Question Answering)was introduced.Secondly,the issues and challenges of visual question answering were discussed,and at the same time,some promising discussion on the particular methodologies was given.Thirdly,the key sub-problems affecting visual question answering were summarized and analyzed.Then,the current commonly used data sets and evaluation indicators were summarized.Next,in view of the popular algorithms and models in VQA research,comparison of the algorithms and models was summarized and listed.Finally,the future development trend and conclusion of visual question answering were prospected.
基金supported by the National Natural Science Foundation of China(61673381)the National Key R&D Program of China(2018AAA0103103)the Science and Technology Development Fund(0024/2018/A1)。
文摘An obstacle perception system for intelligent vehicle is proposed.The proposed system combines the stereo version technique and the deep learning network model,and is applied to obstacle perception tasks in complex environment.In this paper,we provide a complete system design project,which includes the hardware parameters,software framework,algorithm principle,and optimization method.In addition,special experiments are designed to demonstrate that the performance of the proposed system meets the requirements of actual application.The experiment results show that the proposed system is valid to both standard obstacles and non-standard obstacles,and suitable for different weather and lighting conditions in complex environment.It announces that the proposed system is flexible and robust to the intelligent vehicle.
文摘To address the issue of neglecting scenarios involving joint operations and collaborative drone swarm operations in air combat target intent recognition.This paper proposes a transfer learning-based intention prediction model for drone formation targets in air combat.This model recognizes the intentions of multiple aerial targets by extracting spatial features among the targets at each moment.Simulation results demonstrate that,compared to classical intention recognition models,the proposed model in this paper achieves higher accuracy in identifying the intentions of drone swarm targets in air combat scenarios.
基金supported by the Key Research and Development Program of Shaanxi (2022GXLH-02-09)the Aeronautical Science Foundation of China (20200051053001)the Natural Science Basic Research Program of Shaanxi (2020JM-147)。
文摘Autonomous umanned aerial vehicle(UAV) manipulation is necessary for the defense department to execute tactical missions given by commanders in the future unmanned battlefield. A large amount of research has been devoted to improving the autonomous decision-making ability of UAV in an interactive environment, where finding the optimal maneuvering decisionmaking policy became one of the key issues for enabling the intelligence of UAV. In this paper, we propose a maneuvering decision-making algorithm for autonomous air-delivery based on deep reinforcement learning under the guidance of expert experience. Specifically, we refine the guidance towards area and guidance towards specific point tasks for the air-delivery process based on the traditional air-to-surface fire control methods.Moreover, we construct the UAV maneuvering decision-making model based on Markov decision processes(MDPs). Specifically, we present a reward shaping method for the guidance towards area and guidance towards specific point tasks using potential-based function and expert-guided advice. The proposed algorithm could accelerate the convergence of the maneuvering decision-making policy and increase the stability of the policy in terms of the output during the later stage of training process. The effectiveness of the proposed maneuvering decision-making policy is illustrated by the curves of training parameters and extensive experimental results for testing the trained policy.
