Hyperspectral image(HSI)classification is crucial for numerous remote sensing applications.Traditional deep learning methods may miss pixel relationships and context,leading to inefficiencies.This paper introduces the...Hyperspectral image(HSI)classification is crucial for numerous remote sensing applications.Traditional deep learning methods may miss pixel relationships and context,leading to inefficiencies.This paper introduces the spectral band graph convolutional and attention-enhanced CNN joint network(SGCCN),a novel approach that harnesses the power of spectral band graph convolutions for capturing long-range relationships,utilizes local perception of attention-enhanced multi-level convolutions for local spatial feature and employs a dynamic attention mechanism to enhance feature extraction.The SGCCN integrates spectral and spatial features through a self-attention fusion network,significantly improving classification accuracy and efficiency.The proposed method outperforms existing techniques,demonstrating its effectiveness in handling the challenges associated with HSI data.展开更多
Hidden capacity,concealment,security,and robustness are essential indicators of hiding algorithms.Currently,hiding algorithms tend to focus on algorithmic capacity,concealment,and security but often overlook the robus...Hidden capacity,concealment,security,and robustness are essential indicators of hiding algorithms.Currently,hiding algorithms tend to focus on algorithmic capacity,concealment,and security but often overlook the robustness of the algorithms.In practical applications,the container can suffer from damage caused by noise,cropping,and other attacks during transmission,resulting in challenging or even impossible complete recovery of the secret image.An image hiding algorithm based on dynamic region attention in the multi-scale wavelet domain is proposed to address this issue and enhance the robustness of hiding algorithms.In this proposed algorithm,a secret image of size 256×256 is first decomposed using an eight-level Haar wavelet transform.The wavelet transform generates one coefficient in the approximation component and twenty-four detail bands,which are then embedded into the carrier image via a hiding network.During the recovery process,the container image is divided into four non-overlapping parts,each employed to reconstruct a low-resolution secret image.These lowresolution secret images are combined using densemodules to obtain a high-quality secret image.The experimental results showed that even under destructive attacks on the container image,the proposed algorithm is successful in recovering a high-quality secret image,indicating that the algorithm exhibits a high degree of robustness against various attacks.The proposed algorithm effectively addresses the robustness issue by incorporating both spatial and channel attention mechanisms in the multi-scale wavelet domain,making it suitable for practical applications.In conclusion,the image hiding algorithm introduced in this study offers significant improvements in robustness compared to existing algorithms.Its ability to recover high-quality secret images even in the presence of destructive attacksmakes it an attractive option for various applications.Further research and experimentation can explore the algorithm’s performance under different scenarios and expand its potential applications.展开更多
In the production processes of modern industry,accurate assessment of the system’s health state and traceability non-optimal factors are key to ensuring“safe,stable,long-term,full load and optimal”operation of the ...In the production processes of modern industry,accurate assessment of the system’s health state and traceability non-optimal factors are key to ensuring“safe,stable,long-term,full load and optimal”operation of the production process.The benzene-to-ethylene ratio control system is a complex system based on anMPC-PID doublelayer architecture.Taking into consideration the interaction between levels,coupling between loops and conditions of incomplete operation data,this paper proposes a health assessment method for the dual-layer control system by comprehensively utilizing deep learning technology.Firstly,according to the results of the pre-assessment of the system layers and loops bymultivariate statisticalmethods,seven characteristic parameters that have a significant impact on the health state of the system are identified.Next,aiming at the problem of incomplete assessment data set due to the uneven distribution of actual system operating health state,the original unbalanced dataset is augmented using aWasserstein generative adversarial network with gradient penalty term,and a complete dataset is obtained to characterise all the health states of the system.On this basis,a new deep learning-based health assessment framework for the benzeneto-ethylene ratio control system is constructed based on traditionalmultivariate statistical assessment.This framework can overcome the shortcomings of the linear weighted fusion related to the coupling and nonlinearity of the subsystem health state at different layers,and reduce the dependence of the prior knowledge.Furthermore,by introducing a dynamic attention mechanism(AM)into the convolutional neural network(CNN),the assessment model integrating both assessment and traceability is constructed,which can achieve the health assessment and trace the non-optimal factors of the complex control systems with the double-layer architecture.Finally,the effectiveness and superiority of the proposed method have been verified by the benzene-ethylene ratio control system of the alkylation process unit in a styrene plant.展开更多
Accurate crowd counting is a critical challenge for intelligent transportation management,especially in high-density rail transit stations.The existing methods often struggle with multi-scale feature adaptability and ...Accurate crowd counting is a critical challenge for intelligent transportation management,especially in high-density rail transit stations.The existing methods often struggle with multi-scale feature adaptability and dense occlusion robustness in real-world complex settings,leading to reduced accuracy.To address these issues,we propose the multi-level attention network(MLANet),which enhances passenger flow counting and provides essential data to support passenger safety by integrating a multi-scale attention module(MSAM)and a dynamic Gaussian attention module(DGAM).MSAM leverages parallel convolution and spatial attention mechanisms to capture multi-scale crowd features effectively,whereas DGAM dynamically refines attention regions by adapting Gaussian distribution parameters to varying crowd densities.Additionally,by integrating transformer-based global attention,MLANet significantly enhances feature representation in complex crowd environments.To further enhance robustness,we design a hybrid loss function that combines Euclidean loss and dynamic Gaussian attention loss to optimize feature distribution learning.The experimental results demonstrate that MLANet reaches an advanced level on the ShanghaiTech A/B,University of Central Florida(UCF)-QNRF,Johns Hopkins University(JHU)-Crowd++benchmark datasets and the self-built Railway-Station dataset.The proposed approach offers a novel solution for accurate crowd counting in high-density passenger flow environments and holds significant practical value for ensuring operational safety.展开更多
Owing to the continuous barrage of cyber threats,there is a massive amount of cyber threat intelligence.However,a great deal of cyber threat intelligence come from textual sources.For analysis of cyber threat intellig...Owing to the continuous barrage of cyber threats,there is a massive amount of cyber threat intelligence.However,a great deal of cyber threat intelligence come from textual sources.For analysis of cyber threat intelligence,many security analysts rely on cumbersome and time-consuming manual efforts.Cybersecurity knowledge graph plays a significant role in automatics analysis of cyber threat intelligence.As the foundation for constructing cybersecurity knowledge graph,named entity recognition(NER)is required for identifying critical threat-related elements from textual cyber threat intelligence.Recently,deep neural network-based models have attained very good results in NER.However,the performance of these models relies heavily on the amount of labeled data.Since labeled data in cybersecurity is scarce,in this paper,we propose an adversarial active learning framework to effectively select the informative samples for further annotation.In addition,leveraging the long short-term memory(LSTM)network and the bidirectional LSTM(BiLSTM)network,we propose a novel NER model by introducing a dynamic attention mechanism into the BiLSTM-LSTM encoderdecoder.With the selected informative samples annotated,the proposed NER model is retrained.As a result,the performance of the NER model is incrementally enhanced with low labeling cost.Experimental results show the effectiveness of the proposed method.展开更多
Deep neural network-based relational extraction research has made significant progress in recent years,andit provides data support for many natural language processing downstream tasks such as building knowledgegraph,...Deep neural network-based relational extraction research has made significant progress in recent years,andit provides data support for many natural language processing downstream tasks such as building knowledgegraph,sentiment analysis and question-answering systems.However,previous studies ignored much unusedstructural information in sentences that could enhance the performance of the relation extraction task.Moreover,most existing dependency-based models utilize self-attention to distinguish the importance of context,whichhardly deals withmultiple-structure information.To efficiently leverage multiple structure information,this paperproposes a dynamic structure attention mechanism model based on textual structure information,which deeplyintegrates word embedding,named entity recognition labels,part of speech,dependency tree and dependency typeinto a graph convolutional network.Specifically,our model extracts text features of different structures from theinput sentence.Textual Structure information Graph Convolutional Networks employs the dynamic structureattention mechanism to learn multi-structure attention,effectively distinguishing important contextual features invarious structural information.In addition,multi-structure weights are carefully designed as amergingmechanismin the different structure attention to dynamically adjust the final attention.This paper combines these featuresand trains a graph convolutional network for relation extraction.We experiment on supervised relation extractiondatasets including SemEval 2010 Task 8,TACRED,TACREV,and Re-TACED,the result significantly outperformsthe previous.展开更多
The precise and timely extraction of railway signals is crucial for the creation of railway electronic maps.This paper introduces a novel real-time detection approach for dynamically adjusting railway signals,leveragi...The precise and timely extraction of railway signals is crucial for the creation of railway electronic maps.This paper introduces a novel real-time detection approach for dynamically adjusting railway signals,leveraging an enhanced Real-Time DEtection TRansformer(RT-DETR)model.The enhancement involves the integration of a vision Transformer with Dynamically Quantifiable Sampling Attention Mechanism(DQSAM)into the ResNet50 backbone of the RT-DETR framework,thereby enhancing the model’s efficiency and accuracy in handling intricate visual tasks.Secondly,an ultra-lightweight and effective Dynamic Grouping upSampler(DyGSample)is inserted into the efficient hybrid encode module as the up-sampling part.This operator can effectively upsample the feature graph without increasing the computational burden,and improve the model resolution and detail capture ability.In addition,in order to solve the problem of deep layer of model network and high operating cost,a new bounding box similarity loss function of rotation intersection over union based on minimum point distance is adopted in this paper,which takes into account all relevant factors of existing loss functions,namely overlapping or non-overlapping regions,center point distance,width and height deviation,and simplifies the calculation process.As a lightweight signal detection model with ultra-fast,high real-time,and high precision,the detection accuracy of this method is improved from 90.21%to 97.45%,which proves the superior performance and effectiveness of the improved real-time dynamic adjustment RT-DETR model in railway signal extraction.展开更多
Recognizing the intricate spatiotemporal correlation(STC)among wind farms(WFs)is critical to achieving better predictions for wind farm clusters(WFCs).To describe the STC accurately,this paper em-ploys the wind angula...Recognizing the intricate spatiotemporal correlation(STC)among wind farms(WFs)is critical to achieving better predictions for wind farm clusters(WFCs).To describe the STC accurately,this paper em-ploys the wind angular field method to transform the wind series of WFs into different 2-D feature maps,and then construct homogeneous and heterogeneous STC graphs from these maps.The graphs are dynamically updated at the frequency of data update to capture time-varying STC among WFs.Finally,a dynamic graph attention network,designed according to the STC graphs,is established for WFC prediction.Through the above process,dynamic and accurate descriptions of STC are realized in WFC prediction.From the case study of a large-scale WFC with a capacity over 5800 MW in Northeast China,the proposed method reduces the root mean square error of the prediction in the next 24 hours by 2.67%.展开更多
基金supported in part by the National Natural Science Foundations of China(No.61801214)the Postgraduate Research Practice Innovation Program of NUAA(No.xcxjh20231504)。
文摘Hyperspectral image(HSI)classification is crucial for numerous remote sensing applications.Traditional deep learning methods may miss pixel relationships and context,leading to inefficiencies.This paper introduces the spectral band graph convolutional and attention-enhanced CNN joint network(SGCCN),a novel approach that harnesses the power of spectral band graph convolutions for capturing long-range relationships,utilizes local perception of attention-enhanced multi-level convolutions for local spatial feature and employs a dynamic attention mechanism to enhance feature extraction.The SGCCN integrates spectral and spatial features through a self-attention fusion network,significantly improving classification accuracy and efficiency.The proposed method outperforms existing techniques,demonstrating its effectiveness in handling the challenges associated with HSI data.
基金partly supported by the National Natural Science Foundation of China(Jianhua Wu,Grant No.62041106).
文摘Hidden capacity,concealment,security,and robustness are essential indicators of hiding algorithms.Currently,hiding algorithms tend to focus on algorithmic capacity,concealment,and security but often overlook the robustness of the algorithms.In practical applications,the container can suffer from damage caused by noise,cropping,and other attacks during transmission,resulting in challenging or even impossible complete recovery of the secret image.An image hiding algorithm based on dynamic region attention in the multi-scale wavelet domain is proposed to address this issue and enhance the robustness of hiding algorithms.In this proposed algorithm,a secret image of size 256×256 is first decomposed using an eight-level Haar wavelet transform.The wavelet transform generates one coefficient in the approximation component and twenty-four detail bands,which are then embedded into the carrier image via a hiding network.During the recovery process,the container image is divided into four non-overlapping parts,each employed to reconstruct a low-resolution secret image.These lowresolution secret images are combined using densemodules to obtain a high-quality secret image.The experimental results showed that even under destructive attacks on the container image,the proposed algorithm is successful in recovering a high-quality secret image,indicating that the algorithm exhibits a high degree of robustness against various attacks.The proposed algorithm effectively addresses the robustness issue by incorporating both spatial and channel attention mechanisms in the multi-scale wavelet domain,making it suitable for practical applications.In conclusion,the image hiding algorithm introduced in this study offers significant improvements in robustness compared to existing algorithms.Its ability to recover high-quality secret images even in the presence of destructive attacksmakes it an attractive option for various applications.Further research and experimentation can explore the algorithm’s performance under different scenarios and expand its potential applications.
基金supported by the National Science Foundation of China(62263020)the Key Project of Natural Science Foundation of Gansu Province(25JRRA061)+1 种基金the Key R&D Program of Gansu Province(23YFGA0061)the Scientific Research Initiation Fund of Lanzhou University of Technology(061602).
文摘In the production processes of modern industry,accurate assessment of the system’s health state and traceability non-optimal factors are key to ensuring“safe,stable,long-term,full load and optimal”operation of the production process.The benzene-to-ethylene ratio control system is a complex system based on anMPC-PID doublelayer architecture.Taking into consideration the interaction between levels,coupling between loops and conditions of incomplete operation data,this paper proposes a health assessment method for the dual-layer control system by comprehensively utilizing deep learning technology.Firstly,according to the results of the pre-assessment of the system layers and loops bymultivariate statisticalmethods,seven characteristic parameters that have a significant impact on the health state of the system are identified.Next,aiming at the problem of incomplete assessment data set due to the uneven distribution of actual system operating health state,the original unbalanced dataset is augmented using aWasserstein generative adversarial network with gradient penalty term,and a complete dataset is obtained to characterise all the health states of the system.On this basis,a new deep learning-based health assessment framework for the benzeneto-ethylene ratio control system is constructed based on traditionalmultivariate statistical assessment.This framework can overcome the shortcomings of the linear weighted fusion related to the coupling and nonlinearity of the subsystem health state at different layers,and reduce the dependence of the prior knowledge.Furthermore,by introducing a dynamic attention mechanism(AM)into the convolutional neural network(CNN),the assessment model integrating both assessment and traceability is constructed,which can achieve the health assessment and trace the non-optimal factors of the complex control systems with the double-layer architecture.Finally,the effectiveness and superiority of the proposed method have been verified by the benzene-ethylene ratio control system of the alkylation process unit in a styrene plant.
基金supported in part by the National Key R&D Program of China(No.2022YFB4301305).
文摘Accurate crowd counting is a critical challenge for intelligent transportation management,especially in high-density rail transit stations.The existing methods often struggle with multi-scale feature adaptability and dense occlusion robustness in real-world complex settings,leading to reduced accuracy.To address these issues,we propose the multi-level attention network(MLANet),which enhances passenger flow counting and provides essential data to support passenger safety by integrating a multi-scale attention module(MSAM)and a dynamic Gaussian attention module(DGAM).MSAM leverages parallel convolution and spatial attention mechanisms to capture multi-scale crowd features effectively,whereas DGAM dynamically refines attention regions by adapting Gaussian distribution parameters to varying crowd densities.Additionally,by integrating transformer-based global attention,MLANet significantly enhances feature representation in complex crowd environments.To further enhance robustness,we design a hybrid loss function that combines Euclidean loss and dynamic Gaussian attention loss to optimize feature distribution learning.The experimental results demonstrate that MLANet reaches an advanced level on the ShanghaiTech A/B,University of Central Florida(UCF)-QNRF,Johns Hopkins University(JHU)-Crowd++benchmark datasets and the self-built Railway-Station dataset.The proposed approach offers a novel solution for accurate crowd counting in high-density passenger flow environments and holds significant practical value for ensuring operational safety.
基金the National Natural Science Foundation of China undergrant 61501515.
文摘Owing to the continuous barrage of cyber threats,there is a massive amount of cyber threat intelligence.However,a great deal of cyber threat intelligence come from textual sources.For analysis of cyber threat intelligence,many security analysts rely on cumbersome and time-consuming manual efforts.Cybersecurity knowledge graph plays a significant role in automatics analysis of cyber threat intelligence.As the foundation for constructing cybersecurity knowledge graph,named entity recognition(NER)is required for identifying critical threat-related elements from textual cyber threat intelligence.Recently,deep neural network-based models have attained very good results in NER.However,the performance of these models relies heavily on the amount of labeled data.Since labeled data in cybersecurity is scarce,in this paper,we propose an adversarial active learning framework to effectively select the informative samples for further annotation.In addition,leveraging the long short-term memory(LSTM)network and the bidirectional LSTM(BiLSTM)network,we propose a novel NER model by introducing a dynamic attention mechanism into the BiLSTM-LSTM encoderdecoder.With the selected informative samples annotated,the proposed NER model is retrained.As a result,the performance of the NER model is incrementally enhanced with low labeling cost.Experimental results show the effectiveness of the proposed method.
文摘Deep neural network-based relational extraction research has made significant progress in recent years,andit provides data support for many natural language processing downstream tasks such as building knowledgegraph,sentiment analysis and question-answering systems.However,previous studies ignored much unusedstructural information in sentences that could enhance the performance of the relation extraction task.Moreover,most existing dependency-based models utilize self-attention to distinguish the importance of context,whichhardly deals withmultiple-structure information.To efficiently leverage multiple structure information,this paperproposes a dynamic structure attention mechanism model based on textual structure information,which deeplyintegrates word embedding,named entity recognition labels,part of speech,dependency tree and dependency typeinto a graph convolutional network.Specifically,our model extracts text features of different structures from theinput sentence.Textual Structure information Graph Convolutional Networks employs the dynamic structureattention mechanism to learn multi-structure attention,effectively distinguishing important contextual features invarious structural information.In addition,multi-structure weights are carefully designed as amergingmechanismin the different structure attention to dynamically adjust the final attention.This paper combines these featuresand trains a graph convolutional network for relation extraction.We experiment on supervised relation extractiondatasets including SemEval 2010 Task 8,TACRED,TACREV,and Re-TACED,the result significantly outperformsthe previous.
基金supported by the Hunan Provincial Natural Science Foundation of China(Nos.2025JJ70018 and 2025JJ70057)the Hunan Provincial Key Research and Development Program(No.2024JK2065).
文摘The precise and timely extraction of railway signals is crucial for the creation of railway electronic maps.This paper introduces a novel real-time detection approach for dynamically adjusting railway signals,leveraging an enhanced Real-Time DEtection TRansformer(RT-DETR)model.The enhancement involves the integration of a vision Transformer with Dynamically Quantifiable Sampling Attention Mechanism(DQSAM)into the ResNet50 backbone of the RT-DETR framework,thereby enhancing the model’s efficiency and accuracy in handling intricate visual tasks.Secondly,an ultra-lightweight and effective Dynamic Grouping upSampler(DyGSample)is inserted into the efficient hybrid encode module as the up-sampling part.This operator can effectively upsample the feature graph without increasing the computational burden,and improve the model resolution and detail capture ability.In addition,in order to solve the problem of deep layer of model network and high operating cost,a new bounding box similarity loss function of rotation intersection over union based on minimum point distance is adopted in this paper,which takes into account all relevant factors of existing loss functions,namely overlapping or non-overlapping regions,center point distance,width and height deviation,and simplifies the calculation process.As a lightweight signal detection model with ultra-fast,high real-time,and high precision,the detection accuracy of this method is improved from 90.21%to 97.45%,which proves the superior performance and effectiveness of the improved real-time dynamic adjustment RT-DETR model in railway signal extraction.
基金supported by the National Key R&D Program of China(No.2022YFB2403000).
文摘Recognizing the intricate spatiotemporal correlation(STC)among wind farms(WFs)is critical to achieving better predictions for wind farm clusters(WFCs).To describe the STC accurately,this paper em-ploys the wind angular field method to transform the wind series of WFs into different 2-D feature maps,and then construct homogeneous and heterogeneous STC graphs from these maps.The graphs are dynamically updated at the frequency of data update to capture time-varying STC among WFs.Finally,a dynamic graph attention network,designed according to the STC graphs,is established for WFC prediction.Through the above process,dynamic and accurate descriptions of STC are realized in WFC prediction.From the case study of a large-scale WFC with a capacity over 5800 MW in Northeast China,the proposed method reduces the root mean square error of the prediction in the next 24 hours by 2.67%.
