With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based...With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based on GNN can deal with encrypted traffic well. However, existing GNN-based approaches ignore the relationship between client or server packets. In this paper, we design a network traffic topology based on GCN, called Flow Mapping Graph (FMG). FMG establishes sequential edges between vertexes by the arrival order of packets and establishes jump-order edges between vertexes by connecting packets in different bursts with the same direction. It not only reflects the time characteristics of the packet but also strengthens the relationship between the client or server packets. According to FMG, a Traffic Mapping Classification model (TMC-GCN) is designed, which can automatically capture and learn the characteristics and structure information of the top vertex in FMG. The TMC-GCN model is used to classify the encrypted traffic. The encryption stream classification problem is transformed into a graph classification problem, which can effectively deal with data from different data sources and application scenarios. By comparing the performance of TMC-GCN with other classical models in four public datasets, including CICIOT2023, ISCXVPN2016, CICAAGM2017, and GraphDapp, the effectiveness of the FMG algorithm is verified. The experimental results show that the accuracy rate of the TMC-GCN model is 96.13%, the recall rate is 95.04%, and the F1 rate is 94.54%.展开更多
The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and hist...The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and historical context,DL offers a powerful complement by enhancing the speed,objectivity,and precision of the classification process.This research explores the significance of image data augmentation techniques in optimizing the performance of convolutional neural networks(CNNs)for geological image analysis,particularly in the classification of igneous,metamorphic,and sedimentary rock types from rock thin section(RTS)images.This study primarily focuses on classic image augmentation techniques and evaluates their impact on model accuracy and precision.Results demonstrate that augmentation techniques like Equalize significantly enhance the model's classification capabilities,achieving an F1-Score of 0.9869 for igneous rocks,0.9884 for metamorphic rocks,and 0.9929 for sedimentary rocks,representing improvements compared to the baseline original results.Moreover,the weighted average F1-Score across all classes and techniques is 0.9886,indicating an enhancement.Conversely,methods like Distort lead to decreased accuracy and F1-Score,with an F1-Score of 0.949 for igneous rocks,0.954 for metamorphic rocks,and 0.9416 for sedimentary rocks,exacerbating the performance compared to the baseline.The study underscores the practicality of image data augmentation in geological image classification and advocates for the adoption of DL methods in this domain for automation and improved results.The findings of this study can benefit various fields,including remote sensing,mineral exploration,and environmental monitoring,by enhancing the accuracy of geological image analysis both for scientific research and industrial applications.展开更多
In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hi...In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hierarchical attention feature learning,which are used simultaneously to remove noise and retain salient features.In the loss function,it considers the losses of difference in paired images according to the intra-variance and inter-variance.In addition,we also collect the disaster scene dataset from remote sensing images and apply the proposed method to disaster scene classification,which contains complex scenes and multiple types of disasters.Compared to other methods,experimental results show that the DNet with hierarchical attention is robust to different datasets and performs better.展开更多
Network traffic classification is a crucial research area aimed at improving quality of service,simplifying network management,and enhancing network security.To address the growing complexity of cryptography,researche...Network traffic classification is a crucial research area aimed at improving quality of service,simplifying network management,and enhancing network security.To address the growing complexity of cryptography,researchers have proposed various machine learning and deep learning approaches to tackle this challenge.However,existing mainstream methods face several general issues.On one hand,the widely used Transformer architecture exhibits high computational complexity,which negatively impacts its efficiency.On the other hand,traditional methods are often unreliable in traffic representation,frequently losing important byte information while retaining unnecessary biases.To address these problems,this paper introduces the Swin Transformer architecture into the domain of network traffic classification and proposes the NetST(Network Swin Transformer)model.This model improves the Swin Transformer to better accommodate the characteristics of network traffic,effectively addressing efficiency issues.Furthermore,this paper presents a traffic representation scheme designed to extract meaningful information from large volumes of traffic while minimizing bias.We integrate four datasets relevant to network traffic classification for our experiments,and the results demonstrate that NetST achieves a high accuracy rate while maintaining low memory usage.展开更多
Rapid,accurate seed classification of soybean varieties is needed for product quality control.We describe a hyperspectral image-based deep-learning model called Dual Attention Feature Fusion Networks(DAFFnet),which se...Rapid,accurate seed classification of soybean varieties is needed for product quality control.We describe a hyperspectral image-based deep-learning model called Dual Attention Feature Fusion Networks(DAFFnet),which sequentially applies 3D Convolutional Neural Network(CNN)and 2D CNN.A fusion attention mechanism module in 2D CNN permits the model to capture local and global feature information by combining with Convolution Block Attention Module(CBAM)and Mobile Vision Transformer(MViT),outperforming conventional hyperspectral image classification models in seed classification.展开更多
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
Schizophrenia(SZ)stands as a severe psychiatric disorder.This study applied diffusion tensor imaging(DTI)data in conjunction with graph neural networks to distinguish SZ patients from normal controls(NCs)and showcases...Schizophrenia(SZ)stands as a severe psychiatric disorder.This study applied diffusion tensor imaging(DTI)data in conjunction with graph neural networks to distinguish SZ patients from normal controls(NCs)and showcases the superior performance of a graph neural network integrating combined fractional anisotropy and fiber number brain network features,achieving an accuracy of 73.79%in distinguishing SZ patients from NCs.Beyond mere discrimination,our study delved deeper into the advantages of utilizing white matter brain network features for identifying SZ patients through interpretable model analysis and gene expression analysis.These analyses uncovered intricate interrelationships between brain imaging markers and genetic biomarkers,providing novel insights into the neuropathological basis of SZ.In summary,our findings underscore the potential of graph neural networks applied to multimodal DTI data for enhancing SZ detection through an integrated analysis of neuroimaging and genetic features.展开更多
Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not ...Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.展开更多
With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods...With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.展开更多
This study proposes an efficient traffic classification model to address the growing threat of distributed denial-of-service(DDoS)attacks in 5th generation technology standard(5G)slicing networks.The proposed method u...This study proposes an efficient traffic classification model to address the growing threat of distributed denial-of-service(DDoS)attacks in 5th generation technology standard(5G)slicing networks.The proposed method utilizes an ensemble of encoder components from multiple autoencoders to compress and extract latent representations from high-dimensional traffic data.These representations are then used as input for a support vector machine(SVM)-based metadata classifier,enabling precise detection of attack traffic.This architecture is designed to achieve both high detection accuracy and training efficiency,while adapting flexibly to the diverse service requirements and complexity of 5G network slicing.The model was evaluated using the DDoS Datasets 2022,collected in a simulated 5G slicing environment.Experiments were conducted under both class-balanced and class-imbalanced conditions.In the balanced setting,the model achieved an accuracy of 89.33%,an F1-score of 88.23%,and an Area Under the Curve(AUC)of 89.45%.In the imbalanced setting(attack:normal 7:3),the model maintained strong robustness,=achieving a recall of 100%and an F1-score of 90.91%,demonstrating its effectiveness in diverse real-world scenarios.Compared to existing AI-based detection methods,the proposed model showed higher precision,better handling of class imbalance,and strong generalization performance.Moreover,its modular structure is well-suited for deployment in containerized network function(NF)environments,making it a practical solution for real-world 5G infrastructure.These results highlight the potential of the proposed approach to enhance both the security and operational resilience of 5G slicing networks.展开更多
This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophagea...This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.展开更多
Current concrete surface crack detection methods cannot simultaneously achieve high detection accuracy and efficiency.Thus,this study focuses on the recognition and classification of crack images and proposes a concre...Current concrete surface crack detection methods cannot simultaneously achieve high detection accuracy and efficiency.Thus,this study focuses on the recognition and classification of crack images and proposes a concrete crack detection method that integrates the Inception module and a quantum convolutional neural network.First,the features of concrete cracks are highlighted by image gray processing,morphological operations,and threshold segmentation,and then the image is quantum coded by angle coding to transform the classical image information into quantum image information.Then,quantum circuits are used to implement classical image convolution operations to improve the convergence speed of the model and enhance the image representation.Second,two image input paths are designed:one with a quantum convolutional layer and the other with a classical convolutional layer.Finally,comparative experiments are conducted using different parameters to determine the optimal concrete crack classification parameter values for concrete crack image classification.Experimental results show that the method is suitable for crack classification in different scenarios,and training speed is greatly improved compared with that of existing deep learning models.The two evaluation metrics,accuracy and recall,are considerably enhanced.展开更多
In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shi...In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shift and complex multipath channel conditions,extracting discriminative features from high-order modulation signals and ensuring model inter-pretability remain challenging.To address these issues,this paper proposes a Fourier attention net-work(FAttNet),which combines an attention mechanism with a Fourier analysis network(FAN).Specifically,the method directly converts the input signal to the frequency domain using the FAN,thereby obtaining frequency features that reflect the periodic variations in amplitude and phase.Abuilt-in attention mechanism then automatically calculates the weights for each frequency band,focusing on the most discriminative components.This approach improves both classification accu-racy and model interpretability.Experimental validation was conducted via high-order modulationsimulation using an RF testbed.The results show that under three different Doppler frequencyshifts and complex multipath channel conditions,with a signal-to-noise ratio of 10 dB,the classifi-cation accuracy can reach 89.1%,90.4%and 90%,all of which are superior to the current main-stream methods.The proposed approach offers practical value for dynamic spectrum access and sig-nal security detection,and it makes important theoretical contributions to the application of deeplearning in complex electromagnetic signal recognition.展开更多
Physiological signals such as electroencephalogram(EEG)signals are often corrupted by artifacts during the acquisition and processing.Some of these artifacts may deteriorate the essential properties of the signal that...Physiological signals such as electroencephalogram(EEG)signals are often corrupted by artifacts during the acquisition and processing.Some of these artifacts may deteriorate the essential properties of the signal that pertains to meaningful information.Most of these artifacts occur due to the involuntary movements or actions the human does during the acquisition process.So,it is recommended to eliminate these artifacts with signal processing approaches.This paper presents two mechanisms of classification and elimination of artifacts.In the first step,a customized deep network is employed to classify clean EEG signals and artifact-included signals.The classification is performed at the feature level,where common space pattern features are extracted with convolutional layers,and these features are later classified with a support vector machine classifier.In the second stage of the work,the artifact signals are decomposed with empirical mode decomposition,and they are then eliminated with the proposed adaptive thresholding mechanism where the threshold value changes for every intrinsic mode decomposition in the iterative mechanism.展开更多
Leaf disease identification is one of the most promising applications of convolutional neural networks(CNNs).This method represents a significant step towards revolutionizing agriculture by enabling the quick and accu...Leaf disease identification is one of the most promising applications of convolutional neural networks(CNNs).This method represents a significant step towards revolutionizing agriculture by enabling the quick and accurate assessment of plant health.In this study,a CNN model was specifically designed and tested to detect and categorize diseases on fig tree leaves.The researchers utilized a dataset of 3422 images,divided into four classes:healthy,fig rust,fig mosaic,and anthracnose.These diseases can significantly reduce the yield and quality of fig tree fruit.The objective of this research is to develop a CNN that can identify and categorize diseases in fig tree leaves.The data for this study was collected from gardens in the Amandi and Mamash Khail Bannu districts of the Khyber Pakhtunkhwa region in Pakistan.To minimize the risk of overfitting and enhance the model’s performance,early stopping techniques and data augmentation were employed.As a result,the model achieved a training accuracy of 91.53%and a validation accuracy of 90.12%,which are considered respectable.This comprehensive model assists farmers in the early identification and categorization of fig tree leaf diseases.Our experts believe that CNNs could serve as valuable tools for accurate disease classification and detection in precision agriculture.We recommend further research to explore additional data sources and more advanced neural networks to improve the model’s accuracy and applicability.Future research will focus on expanding the dataset by including new diseases and testing the model in real-world scenarios to enhance sustainable farming practices.展开更多
The optimized color space is searched by using the wavelet scattering network in the KTH_TIPS_COL color image database for image texture classification. The effect of choosing the color space on the classification acc...The optimized color space is searched by using the wavelet scattering network in the KTH_TIPS_COL color image database for image texture classification. The effect of choosing the color space on the classification accuracy is investigated by converting red green blue （RGB） color space to various other color spaces. The results show that the classification performance generally changes to a large degree when performing color texture classification in various color spaces, and the opponent RGB-based wavelet scattering network outperforms other color spaces-based wavelet scattering networks. Considering that color spaces can be changed into each other, therefore, when dealing with the problem of color texture classification, converting other color spaces to the opponent RGB color space is recommended before performing the wavelet scattering network.展开更多
A rough set based corner classification neural network, the Rough-CC4, is presented to solve document classification problems such as document representation of different document sizes, document feature selection and...A rough set based corner classification neural network, the Rough-CC4, is presented to solve document classification problems such as document representation of different document sizes, document feature selection and document feature encoding. In the Rough-CC4, the documents are described by the equivalent classes of the approximate words. By this method, the dimensions representing the documents can be reduced, which can solve the precision problems caused by the different document sizes and also blur the differences caused by the approximate words. In the Rough-CC4, a binary encoding method is introduced, through which the importance of documents relative to each equivalent class is encoded. By this encoding method, the precision of the Rough-CC4 is improved greatly and the space complexity of the Rough-CC4 is reduced. The Rough-CC4 can be used in automatic classification of documents.展开更多
In order to reduce amount of data storage and improve processing capacity of the system, this paper proposes a new classification method of data source by combining phase synchronization model in network clusteri...In order to reduce amount of data storage and improve processing capacity of the system, this paper proposes a new classification method of data source by combining phase synchronization model in network clustering with cloud model. Firstly, taking data source as a complex network, after the topography of network is obtained, the cloud model of each node data is determined by fuzzy analytic hierarchy process （AHP）. Secondly, by calculating expectation, entropy and hyper entropy of the cloud model, comprehensive coupling strength is got and then it is regarded as the edge weight of topography. Finally, distribution curve is obtained by iterating the phase of each node by means of phase synchronization model. Thus classification of data source is completed. This method can not only provide convenience for storage, cleaning and compression of data, but also improve the efficiency of data analysis.展开更多
Proper quality planning of limestone raw materials is an essential job of maintaining desired feed in cement plant. Rock-type identification is an integrated part of quality planning for limestone mine. In this paper,...Proper quality planning of limestone raw materials is an essential job of maintaining desired feed in cement plant. Rock-type identification is an integrated part of quality planning for limestone mine. In this paper, a computer vision-based rock-type classification algorithm is proposed for fast and reliable identification without human intervention. A laboratory scale vision-based model was developed using probabilistic neural network（PNN） where color histogram features are used as input. The color image histogram-based features that include weighted mean, skewness and kurtosis features are extracted for all three color space red, green, and blue. A total nine features are used as input for the PNN classification model. The smoothing parameter for PNN model is selected judicially to develop an optimal or close to the optimum classification model. The developed PPN is validated using the test data set and results reveal that the proposed vision-based model can perform satisfactorily for classifying limestone rocktypes. Overall the error of mis-classification is below 6%. When compared with other three classification algorithms, it is observed that the proposed method performs substantially better than all three classification algorithms.展开更多
A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a force...A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a forced aligned Hidden Markov Model(HMM) state sequence obtained from the GMM-based acoustic model. Therefore, it requires a long computation time for training both the GMM-based acoustic model and a deep learning-based acoustic model. In order to solve this problem, an acoustic model using CTC algorithm is proposed. CTC algorithm does not require the GMM-based acoustic model because it does not use the forced aligned HMM state sequence. However, previous works on a LSTM RNN-based acoustic model using CTC used a small-scale training corpus. In this paper, the LSTM RNN-based acoustic model using CTC is trained on a large-scale training corpus and its performance is evaluated. The implemented acoustic model has a performance of 6.18% and 15.01% in terms of Word Error Rate(WER) for clean speech and noisy speech, respectively. This is similar to a performance of the acoustic model based on the hybrid method.展开更多
基金supported by the National Key Research and Development Program of China No.2023YFA1009500.
文摘With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based on GNN can deal with encrypted traffic well. However, existing GNN-based approaches ignore the relationship between client or server packets. In this paper, we design a network traffic topology based on GCN, called Flow Mapping Graph (FMG). FMG establishes sequential edges between vertexes by the arrival order of packets and establishes jump-order edges between vertexes by connecting packets in different bursts with the same direction. It not only reflects the time characteristics of the packet but also strengthens the relationship between the client or server packets. According to FMG, a Traffic Mapping Classification model (TMC-GCN) is designed, which can automatically capture and learn the characteristics and structure information of the top vertex in FMG. The TMC-GCN model is used to classify the encrypted traffic. The encryption stream classification problem is transformed into a graph classification problem, which can effectively deal with data from different data sources and application scenarios. By comparing the performance of TMC-GCN with other classical models in four public datasets, including CICIOT2023, ISCXVPN2016, CICAAGM2017, and GraphDapp, the effectiveness of the FMG algorithm is verified. The experimental results show that the accuracy rate of the TMC-GCN model is 96.13%, the recall rate is 95.04%, and the F1 rate is 94.54%.
文摘The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and historical context,DL offers a powerful complement by enhancing the speed,objectivity,and precision of the classification process.This research explores the significance of image data augmentation techniques in optimizing the performance of convolutional neural networks(CNNs)for geological image analysis,particularly in the classification of igneous,metamorphic,and sedimentary rock types from rock thin section(RTS)images.This study primarily focuses on classic image augmentation techniques and evaluates their impact on model accuracy and precision.Results demonstrate that augmentation techniques like Equalize significantly enhance the model's classification capabilities,achieving an F1-Score of 0.9869 for igneous rocks,0.9884 for metamorphic rocks,and 0.9929 for sedimentary rocks,representing improvements compared to the baseline original results.Moreover,the weighted average F1-Score across all classes and techniques is 0.9886,indicating an enhancement.Conversely,methods like Distort lead to decreased accuracy and F1-Score,with an F1-Score of 0.949 for igneous rocks,0.954 for metamorphic rocks,and 0.9416 for sedimentary rocks,exacerbating the performance compared to the baseline.The study underscores the practicality of image data augmentation in geological image classification and advocates for the adoption of DL methods in this domain for automation and improved results.The findings of this study can benefit various fields,including remote sensing,mineral exploration,and environmental monitoring,by enhancing the accuracy of geological image analysis both for scientific research and industrial applications.
基金Supported by the National Natural Science Foundation of China(61601176)。
文摘In this paper,we propose hierarchical attention dual network(DNet)for fine-grained image classification.The DNet can randomly select pairs of inputs from the dataset and compare the differences between them through hierarchical attention feature learning,which are used simultaneously to remove noise and retain salient features.In the loss function,it considers the losses of difference in paired images according to the intra-variance and inter-variance.In addition,we also collect the disaster scene dataset from remote sensing images and apply the proposed method to disaster scene classification,which contains complex scenes and multiple types of disasters.Compared to other methods,experimental results show that the DNet with hierarchical attention is robust to different datasets and performs better.
基金supported by National Natural Science Foundation of China(62473341)Key Technologies R&D Program of Henan Province(242102211071,252102211086,252102210166).
文摘Network traffic classification is a crucial research area aimed at improving quality of service,simplifying network management,and enhancing network security.To address the growing complexity of cryptography,researchers have proposed various machine learning and deep learning approaches to tackle this challenge.However,existing mainstream methods face several general issues.On one hand,the widely used Transformer architecture exhibits high computational complexity,which negatively impacts its efficiency.On the other hand,traditional methods are often unreliable in traffic representation,frequently losing important byte information while retaining unnecessary biases.To address these problems,this paper introduces the Swin Transformer architecture into the domain of network traffic classification and proposes the NetST(Network Swin Transformer)model.This model improves the Swin Transformer to better accommodate the characteristics of network traffic,effectively addressing efficiency issues.Furthermore,this paper presents a traffic representation scheme designed to extract meaningful information from large volumes of traffic while minimizing bias.We integrate four datasets relevant to network traffic classification for our experiments,and the results demonstrate that NetST achieves a high accuracy rate while maintaining low memory usage.
基金supported by Natural Science Foundation of Heilongjiang Province of China(SS2021C005)Province Key Research and Development Program of Heilongjiang Province of China(GZ20220121)the Agricultural Science and Technology Innovation Program of the Chinese Academy of Agricultural Sciences.
文摘Rapid,accurate seed classification of soybean varieties is needed for product quality control.We describe a hyperspectral image-based deep-learning model called Dual Attention Feature Fusion Networks(DAFFnet),which sequentially applies 3D Convolutional Neural Network(CNN)and 2D CNN.A fusion attention mechanism module in 2D CNN permits the model to capture local and global feature information by combining with Convolution Block Attention Module(CBAM)and Mobile Vision Transformer(MViT),outperforming conventional hyperspectral image classification models in seed classification.
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supported by the National Natural Science Foundation of China(62276049,61701078,61872068,and 62006038)the Natural Science Foundation of Sichuan Province(2025ZNSFSC0487)+3 种基金the Science and Technology Innovation 2030-Brain Science and Brain-Inspired Intelligence Project(2021ZD0200200)the National Key R&D Program of China(2023YFE0118600)Sichuan Province Science and Technology Support Program(2019YJ0193,2021YFG0126,2021YFG0366,and 2022YFS0180)Medico-Engineering Cooperation Funds from the University of Electronic Science and Technology of China(ZYGX2021YGLH014).
文摘Schizophrenia(SZ)stands as a severe psychiatric disorder.This study applied diffusion tensor imaging(DTI)data in conjunction with graph neural networks to distinguish SZ patients from normal controls(NCs)and showcases the superior performance of a graph neural network integrating combined fractional anisotropy and fiber number brain network features,achieving an accuracy of 73.79%in distinguishing SZ patients from NCs.Beyond mere discrimination,our study delved deeper into the advantages of utilizing white matter brain network features for identifying SZ patients through interpretable model analysis and gene expression analysis.These analyses uncovered intricate interrelationships between brain imaging markers and genetic biomarkers,providing novel insights into the neuropathological basis of SZ.In summary,our findings underscore the potential of graph neural networks applied to multimodal DTI data for enhancing SZ detection through an integrated analysis of neuroimaging and genetic features.
基金supported by the National Natural Science Foundation of China(61962016)the Ministry of Science and Technology of China(G2022033002L)+1 种基金National Natural Science Foundation of Guangxi(2022JJA170057)Guangxi Education Department’s Project on Improving the Basic Research Ability of Young and Middleaged Teachers in Universities(2023ky0812,Research on Statistical Network Delay Predictions in Large-scale SDNs).
文摘Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.
基金supported by the National Key Research and Development Program of China No.2023YFB2705000.
文摘With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.
基金supported by an Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korean government(MSIT)(RS-2024-00438156,Development of Security Resilience Technology Based on Network Slicing Services in a 5G Specialized Network).
文摘This study proposes an efficient traffic classification model to address the growing threat of distributed denial-of-service(DDoS)attacks in 5th generation technology standard(5G)slicing networks.The proposed method utilizes an ensemble of encoder components from multiple autoencoders to compress and extract latent representations from high-dimensional traffic data.These representations are then used as input for a support vector machine(SVM)-based metadata classifier,enabling precise detection of attack traffic.This architecture is designed to achieve both high detection accuracy and training efficiency,while adapting flexibly to the diverse service requirements and complexity of 5G network slicing.The model was evaluated using the DDoS Datasets 2022,collected in a simulated 5G slicing environment.Experiments were conducted under both class-balanced and class-imbalanced conditions.In the balanced setting,the model achieved an accuracy of 89.33%,an F1-score of 88.23%,and an Area Under the Curve(AUC)of 89.45%.In the imbalanced setting(attack:normal 7:3),the model maintained strong robustness,=achieving a recall of 100%and an F1-score of 90.91%,demonstrating its effectiveness in diverse real-world scenarios.Compared to existing AI-based detection methods,the proposed model showed higher precision,better handling of class imbalance,and strong generalization performance.Moreover,its modular structure is well-suited for deployment in containerized network function(NF)environments,making it a practical solution for real-world 5G infrastructure.These results highlight the potential of the proposed approach to enhance both the security and operational resilience of 5G slicing networks.
基金funded by Vietnam National Foundation for Science and Technology Development(NAFOSTED)under grant number:NCUD.02-2024.11.
文摘This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.
基金supported by 2023 National College Students'Innovation and Entrepreneurship Training Program project"Building Crack Structure Safety Detection based on Quantum Convolutional Neural Network intelligent Algorithm-A case study of Sanzhuang Town,Donggang District,Rizhao City"(NO.202310429224).
文摘Current concrete surface crack detection methods cannot simultaneously achieve high detection accuracy and efficiency.Thus,this study focuses on the recognition and classification of crack images and proposes a concrete crack detection method that integrates the Inception module and a quantum convolutional neural network.First,the features of concrete cracks are highlighted by image gray processing,morphological operations,and threshold segmentation,and then the image is quantum coded by angle coding to transform the classical image information into quantum image information.Then,quantum circuits are used to implement classical image convolution operations to improve the convergence speed of the model and enhance the image representation.Second,two image input paths are designed:one with a quantum convolutional layer and the other with a classical convolutional layer.Finally,comparative experiments are conducted using different parameters to determine the optimal concrete crack classification parameter values for concrete crack image classification.Experimental results show that the method is suitable for crack classification in different scenarios,and training speed is greatly improved compared with that of existing deep learning models.The two evaluation metrics,accuracy and recall,are considerably enhanced.
基金supported by the National Natural Science Foundation of China(No.62027801).
文摘In modern wireless communication and electromagnetic control,automatic modulationclassification(AMC)of orthogonal frequency division multiplexing(OFDM)signals plays animportant role.However,under Doppler frequency shift and complex multipath channel conditions,extracting discriminative features from high-order modulation signals and ensuring model inter-pretability remain challenging.To address these issues,this paper proposes a Fourier attention net-work(FAttNet),which combines an attention mechanism with a Fourier analysis network(FAN).Specifically,the method directly converts the input signal to the frequency domain using the FAN,thereby obtaining frequency features that reflect the periodic variations in amplitude and phase.Abuilt-in attention mechanism then automatically calculates the weights for each frequency band,focusing on the most discriminative components.This approach improves both classification accu-racy and model interpretability.Experimental validation was conducted via high-order modulationsimulation using an RF testbed.The results show that under three different Doppler frequencyshifts and complex multipath channel conditions,with a signal-to-noise ratio of 10 dB,the classifi-cation accuracy can reach 89.1%,90.4%and 90%,all of which are superior to the current main-stream methods.The proposed approach offers practical value for dynamic spectrum access and sig-nal security detection,and it makes important theoretical contributions to the application of deeplearning in complex electromagnetic signal recognition.
文摘Physiological signals such as electroencephalogram(EEG)signals are often corrupted by artifacts during the acquisition and processing.Some of these artifacts may deteriorate the essential properties of the signal that pertains to meaningful information.Most of these artifacts occur due to the involuntary movements or actions the human does during the acquisition process.So,it is recommended to eliminate these artifacts with signal processing approaches.This paper presents two mechanisms of classification and elimination of artifacts.In the first step,a customized deep network is employed to classify clean EEG signals and artifact-included signals.The classification is performed at the feature level,where common space pattern features are extracted with convolutional layers,and these features are later classified with a support vector machine classifier.In the second stage of the work,the artifact signals are decomposed with empirical mode decomposition,and they are then eliminated with the proposed adaptive thresholding mechanism where the threshold value changes for every intrinsic mode decomposition in the iterative mechanism.
基金the Deanship of Graduate Studies and Scientific Research at Qassim University for financial support(QU-APC-2025).
文摘Leaf disease identification is one of the most promising applications of convolutional neural networks(CNNs).This method represents a significant step towards revolutionizing agriculture by enabling the quick and accurate assessment of plant health.In this study,a CNN model was specifically designed and tested to detect and categorize diseases on fig tree leaves.The researchers utilized a dataset of 3422 images,divided into four classes:healthy,fig rust,fig mosaic,and anthracnose.These diseases can significantly reduce the yield and quality of fig tree fruit.The objective of this research is to develop a CNN that can identify and categorize diseases in fig tree leaves.The data for this study was collected from gardens in the Amandi and Mamash Khail Bannu districts of the Khyber Pakhtunkhwa region in Pakistan.To minimize the risk of overfitting and enhance the model’s performance,early stopping techniques and data augmentation were employed.As a result,the model achieved a training accuracy of 91.53%and a validation accuracy of 90.12%,which are considered respectable.This comprehensive model assists farmers in the early identification and categorization of fig tree leaf diseases.Our experts believe that CNNs could serve as valuable tools for accurate disease classification and detection in precision agriculture.We recommend further research to explore additional data sources and more advanced neural networks to improve the model’s accuracy and applicability.Future research will focus on expanding the dataset by including new diseases and testing the model in real-world scenarios to enhance sustainable farming practices.
基金The National Basic Research Program of China(No.2011CB707904)the National Natural Science Foundation of China(No.61201344,61271312,11301074)+1 种基金the Natural Science Foundation of Jiangsu Province(No.BK2012329)the Specialized Research Fund for the Doctoral Program of Higher Education(No.20110092110023,20120092120036)
文摘The optimized color space is searched by using the wavelet scattering network in the KTH_TIPS_COL color image database for image texture classification. The effect of choosing the color space on the classification accuracy is investigated by converting red green blue （RGB） color space to various other color spaces. The results show that the classification performance generally changes to a large degree when performing color texture classification in various color spaces, and the opponent RGB-based wavelet scattering network outperforms other color spaces-based wavelet scattering networks. Considering that color spaces can be changed into each other, therefore, when dealing with the problem of color texture classification, converting other color spaces to the opponent RGB color space is recommended before performing the wavelet scattering network.
基金The National Natural Science Foundation of China(No.60503020,60373066,60403016,60425206),the Natural Science Foundation of Jiangsu Higher Education Institutions ( No.04KJB520096),the Doctoral Foundation of Nanjing University of Posts and Telecommunication (No.0302).
文摘A rough set based corner classification neural network, the Rough-CC4, is presented to solve document classification problems such as document representation of different document sizes, document feature selection and document feature encoding. In the Rough-CC4, the documents are described by the equivalent classes of the approximate words. By this method, the dimensions representing the documents can be reduced, which can solve the precision problems caused by the different document sizes and also blur the differences caused by the approximate words. In the Rough-CC4, a binary encoding method is introduced, through which the importance of documents relative to each equivalent class is encoded. By this encoding method, the precision of the Rough-CC4 is improved greatly and the space complexity of the Rough-CC4 is reduced. The Rough-CC4 can be used in automatic classification of documents.
基金National Natural Science Foundation of China(No.61171057,No.61503345)Science Foundation for North University of China(No.110246)+1 种基金Specialized Research Fund for Doctoral Program of Higher Education of China(No.20121420110004)International Office of Shanxi Province Education Department of China,and Basic Research Project in Shanxi Province(Young Foundation)
文摘In order to reduce amount of data storage and improve processing capacity of the system, this paper proposes a new classification method of data source by combining phase synchronization model in network clustering with cloud model. Firstly, taking data source as a complex network, after the topography of network is obtained, the cloud model of each node data is determined by fuzzy analytic hierarchy process （AHP）. Secondly, by calculating expectation, entropy and hyper entropy of the cloud model, comprehensive coupling strength is got and then it is regarded as the edge weight of topography. Finally, distribution curve is obtained by iterating the phase of each node by means of phase synchronization model. Thus classification of data source is completed. This method can not only provide convenience for storage, cleaning and compression of data, but also improve the efficiency of data analysis.
文摘Proper quality planning of limestone raw materials is an essential job of maintaining desired feed in cement plant. Rock-type identification is an integrated part of quality planning for limestone mine. In this paper, a computer vision-based rock-type classification algorithm is proposed for fast and reliable identification without human intervention. A laboratory scale vision-based model was developed using probabilistic neural network（PNN） where color histogram features are used as input. The color image histogram-based features that include weighted mean, skewness and kurtosis features are extracted for all three color space red, green, and blue. A total nine features are used as input for the PNN classification model. The smoothing parameter for PNN model is selected judicially to develop an optimal or close to the optimum classification model. The developed PPN is validated using the test data set and results reveal that the proposed vision-based model can perform satisfactorily for classifying limestone rocktypes. Overall the error of mis-classification is below 6%. When compared with other three classification algorithms, it is observed that the proposed method performs substantially better than all three classification algorithms.
基金supported by the Ministry of Trade,Industry & Energy(MOTIE,Korea) under Industrial Technology Innovation Program (No.10063424,'development of distant speech recognition and multi-task dialog processing technologies for in-door conversational robots')
文摘A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a forced aligned Hidden Markov Model(HMM) state sequence obtained from the GMM-based acoustic model. Therefore, it requires a long computation time for training both the GMM-based acoustic model and a deep learning-based acoustic model. In order to solve this problem, an acoustic model using CTC algorithm is proposed. CTC algorithm does not require the GMM-based acoustic model because it does not use the forced aligned HMM state sequence. However, previous works on a LSTM RNN-based acoustic model using CTC used a small-scale training corpus. In this paper, the LSTM RNN-based acoustic model using CTC is trained on a large-scale training corpus and its performance is evaluated. The implemented acoustic model has a performance of 6.18% and 15.01% in terms of Word Error Rate(WER) for clean speech and noisy speech, respectively. This is similar to a performance of the acoustic model based on the hybrid method.
