Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely us...Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely used in pulmonary disease diagnosis,such as pneumonia and tuberculosis.However,traditional feature fusion methods often suffer from feature disparity,information loss,redundancy,and increased complexity,hindering the further extension of DL algorithms.To solve this problem,we propose a Graph-Convolution Fusion Network with Self-Supervised Feature Alignment(Self-FAGCFN)to address the limitations of traditional feature fusion methods in deep learning-based medical image classification for respiratory diseases such as pneumonia and tuberculosis.The network integrates Convolutional Neural Networks(CNNs)for robust feature extraction from two-dimensional grid structures and Graph Convolutional Networks(GCNs)within a Graph Neural Network branch to capture features based on graph structure,focusing on significant node representations.Additionally,an Attention-Embedding Ensemble Block is included to capture critical features from GCN outputs.To ensure effective feature alignment between pre-and post-fusion stages,we introduce a feature alignment loss that minimizes disparities.Moreover,to address the limitations of proposed methods,such as inappropriate centroid discrepancies during feature alignment and class imbalance in the dataset,we develop a Feature-Centroid Fusion(FCF)strategy and a Multi-Level Feature-Centroid Update(MLFCU)algorithm,respectively.Extensive experiments on public datasets LungVision and Chest-Xray demonstrate that the Self-FAGCFN model significantly outperforms existing methods in diagnosing pneumonia and tuberculosis,highlighting its potential for practical medical applications.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
Wireless sensor network deployment optimization is a classic NP-hard problem and a popular topic in academic research.However,the current research on wireless sensor network deployment problems uses overly simplistic ...Wireless sensor network deployment optimization is a classic NP-hard problem and a popular topic in academic research.However,the current research on wireless sensor network deployment problems uses overly simplistic models,and there is a significant gap between the research results and actual wireless sensor networks.Some scholars have now modeled data fusion networks to make them more suitable for practical applications.This paper will explore the deployment problem of a stochastic data fusion wireless sensor network(SDFWSN),a model that reflects the randomness of environmental monitoring and uses data fusion techniques widely used in actual sensor networks for information collection.The deployment problem of SDFWSN is modeled as a multi-objective optimization problem.The network life cycle,spatiotemporal coverage,detection rate,and false alarm rate of SDFWSN are used as optimization objectives to optimize the deployment of network nodes.This paper proposes an enhanced multi-objective mongoose optimization algorithm(EMODMOA)to solve the deployment problem of SDFWSN.First,to overcome the shortcomings of the DMOA algorithm,such as its low convergence and tendency to get stuck in a local optimum,an encircling and hunting strategy is introduced into the original algorithm to propose the EDMOA algorithm.The EDMOA algorithm is designed as the EMODMOA algorithm by selecting reference points using the K-Nearest Neighbor(KNN)algorithm.To verify the effectiveness of the proposed algorithm,the EMODMOA algorithm was tested at CEC 2020 and achieved good results.In the SDFWSN deployment problem,the algorithm was compared with the Non-dominated Sorting Genetic Algorithm II(NSGAII),Multiple Objective Particle Swarm Optimization(MOPSO),Multi-Objective Evolutionary Algorithm based on Decomposition(MOEA/D),and Multi-Objective Grey Wolf Optimizer(MOGWO).By comparing and analyzing the performance evaluation metrics and optimization results of the objective functions of the multi-objective algorithms,the algorithm outperforms the other algorithms in the SDFWSN deployment results.To better demonstrate the superiority of the algorithm,simulations of diverse test cases were also performed,and good results were obtained.展开更多
A machine learning approach based on Bayesian neural networks was developed to predict the complete fusion cross-sections of weakly bound nuclei.This method was trained and validated using 475 experimental data points...A machine learning approach based on Bayesian neural networks was developed to predict the complete fusion cross-sections of weakly bound nuclei.This method was trained and validated using 475 experimental data points from 39 reaction systems induced by ^(6,7)Li,^(9)Be,and ^(10)B.The constructed Bayesian neural network demonstrated a high degree of accuracy in evaluating complete fusion cross-sections.By comparing the predicted cross-sections with those obtained from a single-barrier penetration model,the suppression effect of ^(6,7)Li and ^(9)Be with a stable nucleus was systematically analyzed.In the cases of ^(6)Li and ^(7)Li,less suppression was predicted for relatively light-mass targets than for heavy-mass targets,and a notably distinct dependence relationship was identified,suggesting that the predominant breakup mechanisms might change in different mass target regions.In addition,minimum suppression factors were predicted to occur near target nuclei with neutron-closed shell.展开更多
Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused inform...Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused information in a single image.One of the critical clinical applications of medical image fusion is to fuse anatomical and functional modalities for rapid diagnosis of malignant tissues.This paper proposes a multimodal medical image fusion network(MMIF-Net)based on multiscale hybrid attention.The method first decomposes the original image to obtain the low-rank and significant parts.Then,to utilize the features at different scales,we add amultiscalemechanism that uses three filters of different sizes to extract the features in the encoded network.Also,a hybrid attention module is introduced to obtain more image details.Finally,the fused images are reconstructed by decoding the network.We conducted experiments with clinical images from brain computed tomography/magnetic resonance.The experimental results show that the multimodal medical image fusion network method based on multiscale hybrid attention works better than other advanced fusion methods.展开更多
Speech Emotion Recognition(SER)has received widespread attention as a crucial way for understanding human emotional states.However,the impact of irrelevant information on speech signals and data sparsity limit the dev...Speech Emotion Recognition(SER)has received widespread attention as a crucial way for understanding human emotional states.However,the impact of irrelevant information on speech signals and data sparsity limit the development of SER system.To address these issues,this paper proposes a framework that incorporates the Attentive Mask Residual Network(AM-ResNet)and the self-supervised learning model Wav2vec 2.0 to obtain AM-ResNet features and Wav2vec 2.0 features respectively,together with a cross-attention module to interact and fuse these two features.The AM-ResNet branch mainly consists of maximum amplitude difference detection,mask residual block,and an attention mechanism.Among them,the maximum amplitude difference detection and the mask residual block act on the pre-processing and the network,respectively,to reduce the impact of silent frames,and the attention mechanism assigns different weights to unvoiced and voiced speech to reduce redundant emotional information caused by unvoiced speech.In the Wav2vec 2.0 branch,this model is introduced as a feature extractor to obtain general speech features(Wav2vec 2.0 features)through pre-training with a large amount of unlabeled speech data,which can assist the SER task and cope with data sparsity problems.In the cross-attention module,AM-ResNet features and Wav2vec 2.0 features are interacted with and fused to obtain the cross-fused features,which are used to predict the final emotion.Furthermore,multi-label learning is also used to add ambiguous emotion utterances to deal with data limitations.Finally,experimental results illustrate the usefulness and superiority of our proposed framework over existing state-of-the-art approaches.展开更多
Skin cancer is the most prevalent cancer globally,primarily due to extensive exposure to Ultraviolet(UV)radiation.Early identification of skin cancer enhances the likelihood of effective treatment,as delays may lead t...Skin cancer is the most prevalent cancer globally,primarily due to extensive exposure to Ultraviolet(UV)radiation.Early identification of skin cancer enhances the likelihood of effective treatment,as delays may lead to severe tumor advancement.This study proposes a novel hybrid deep learning strategy to address the complex issue of skin cancer diagnosis,with an architecture that integrates a Vision Transformer,a bespoke convolutional neural network(CNN),and an Xception module.They were evaluated using two benchmark datasets,HAM10000 and Skin Cancer ISIC.On the HAM10000,the model achieves a precision of 95.46%,an accuracy of 96.74%,a recall of 96.27%,specificity of 96.00%and an F1-Score of 95.86%.It obtains an accuracy of 93.19%,a precision of 93.25%,a recall of 92.80%,a specificity of 92.89%and an F1-Score of 93.19%on the Skin Cancer ISIC dataset.The findings demonstrate that the model that was proposed is robust and trustworthy when it comes to the classification of skin lesions.In addition,the utilization of Explainable AI techniques,such as Grad-CAM visualizations,assists in highlighting the most significant lesion areas that have an impact on the decisions that are made by the model.展开更多
The accurate state of health(SOH)estimation of lithium-ion batteries is crucial for efficient,healthy,and safe operation of battery systems.Extracting meaningful aging information from highly stochastic and noisy data...The accurate state of health(SOH)estimation of lithium-ion batteries is crucial for efficient,healthy,and safe operation of battery systems.Extracting meaningful aging information from highly stochastic and noisy data segments while designing SOH estimation algorithms that efficiently handle the large-scale computational demands of cloud-based battery management systems presents a substantial challenge.In this work,we propose a quantum convolutional neural network(QCNN)model designed for accurate,robust,and generalizable SOH estimation with minimal data and parameter requirements and is compatible with quantum computing cloud platforms in the Noisy Intermediate-Scale Quantum.First,we utilize data from 4 datasets comprising 272 cells,covering 5 chemical compositions,4 rated parameters,and 73operating conditions.We design 5 voltage windows as small as 0.3 V for each cell from incremental capacity peaks for stochastic SOH estimation scenarios generation.We extract 3 effective health indicators(HIs)sequences and develop an automated feature fusion method using quantum rotation gate encoding,achieving an R2of 96%.Subsequently,we design a QCNN whose convolutional layer,constructed with variational quantum circuits,comprises merely 39 parameters.Additionally,we explore the impact of training set size,using strategies,and battery materials on the model’s accuracy.Finally,the QCNN with quantum convolutional layers reduces root mean squared error by 28% and achieves an R^(2)exceeding 96% compared to other three commonly used algorithms.This work demonstrates the effectiveness of quantum encoding for automated feature fusion of HIs extracted from limited discharge data.It highlights the potential of QCNN in improving the accuracy,robustness,and generalization of SOH estimation while dealing with stochastic and noisy data with few parameters and simple structure.It also suggests a new paradigm for leveraging quantum computational power in SOH estimation.展开更多
3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safe...3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safer and faster,poses challenges for accurate volumetric reconstruction due to limited spatial information.This study proposes a 3D reconstruction neural network based on adaptive weight fusion(AdapFusionNet)to achieve high-quality 3D medical image reconstruction from sparse-angle X-ray images.To address the issue of spatial inconsistency in multi-angle image reconstruction,an innovative adaptive fusion module was designed to score initial reconstruction results during the inference stage and perform weighted fusion,thereby improving the final reconstruction quality.The reconstruction network is built on an autoencoder(AE)framework and uses orthogonal-angle X-ray images(frontal and lateral projections)as inputs.The encoder extracts 2D features,which the decoder maps into 3D space.This study utilizes a lung CT dataset to obtain complete three-dimensional volumetric data,from which digitally reconstructed radiographs(DRR)are generated at various angles to simulate X-ray images.Since real-world clinical X-ray images rarely come with perfectly corresponding 3D“ground truth,”using CT scans as the three-dimensional reference effectively supports the training and evaluation of deep networks for sparse-angle X-ray 3D reconstruction.Experiments conducted on the LIDC-IDRI dataset with simulated X-ray images(DRR images)as training data demonstrate the superior performance of AdapFusionNet compared to other fusion methods.Quantitative results show that AdapFusionNet achieves SSIM,PSNR,and MAE values of 0.332,13.404,and 0.163,respectively,outperforming other methods(SingleViewNet:0.289,12.363,0.182;AvgFusionNet:0.306,13.384,0.159).Qualitative analysis further confirms that AdapFusionNet significantly enhances the reconstruction of lung and chest contours while effectively reducing noise during the reconstruction process.The findings demonstrate that AdapFusionNet offers significant advantages in 3D reconstruction of sparse-angle X-ray images.展开更多
An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extra...An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extraction,which are commonly faced by rolling bearings and lead to low diagnostic accuracy.Initially,dual models of the Wasserstein deep convolutional generative adversarial network incorporating gradient penalty(1D-2DWDCGAN)are constructed to augment the original dataset.A self-adaptive loss threshold control training strategy is introduced,and establishing a self-adaptive balancing mechanism for stable model training.Subsequently,a diagnostic model based on multidimensional feature fusion is designed,wherein complex features from various dimensions are extracted,merging the original signal waveform features,structured features,and time-frequency features into a deep composite feature representation that encompasses multiple dimensions and scales;thus,efficient and accurate small sample fault diagnosis is facilitated.Finally,an experiment between the bearing fault dataset of CaseWestern ReserveUniversity and the fault simulation experimental platformdataset of this research group shows that this method effectively supplements the dataset and remarkably improves the diagnostic accuracy.The diagnostic accuracy after data augmentation reached 99.94%and 99.87%in two different experimental environments,respectively.In addition,robustness analysis is conducted on the diagnostic accuracy of the proposed method under different noise backgrounds,verifying its good generalization performance.展开更多
A constructive-pruning hybrid method （CPHM） for radial basis function （RBF） networks is proposed to improve the prediction accuracy of ash fusion temperatures （AFT）. The CPHM incorporates the advantages of the c...A constructive-pruning hybrid method （CPHM） for radial basis function （RBF） networks is proposed to improve the prediction accuracy of ash fusion temperatures （AFT）. The CPHM incorporates the advantages of the construction algorithm and the pruning algorithm of neural networks, and the training process of the CPHM is divided into two stages： rough tuning and fine tuning. In rough tuning, new hidden units are added to the current network until some performance index is satisfied. In fine tuning, the network structure and the model parameters are further adjusted. And, based on components of coal ash, a model using the CPHM is established to predict the AFT. The results show that the CPHM prediction model is characterized by its high precision, compact network structure, as well as strong generalization ability and robustness.展开更多
The spaceborne precipitation radar onboard the Tropical Rainfall Measuring Mission satellite （TRMM PR） can provide good measurement of the vertical structure of reflectivity, while ground radar （GR） has a relative...The spaceborne precipitation radar onboard the Tropical Rainfall Measuring Mission satellite （TRMM PR） can provide good measurement of the vertical structure of reflectivity, while ground radar （GR） has a relatively high horizontal resolution and greater sensitivity. Fusion of TRMM PR and GR reflectivity data may maximize the advantages from both instruments. In this paper, TRMM PR and GR reflectivity data are fused using a neural network （NN）-based approach. The main steps included are： quality control of TRMM PR and GR reflectivity data; spatiotemporal matchup; GR calibration bias correction; conversion of TRMM PR data from Ku to S band; fusion of TRMM PR and GR reflectivity data with an NN method： interpolation of reflectivity data that are below PR＇s sensitivity; blind areas compensation with a distance weighting-based merging approach; combination of three types of data： data with the NN method, data below PR＇s sensitivity and data within compensated blind areas. During the NN fusion step, the TRMM PR data are taken as targets of the training NNs, and gridded GR data after horizontal downsampling at different heights are used as the input. The trained NNs are then used to obtain 3D high-resolution reflectivity from the original GR gridded data. After 3D fusion of the TRMM PR and GR reflectivity data, a more complete and finer-scale 3D radar reflectivity dataset incorporating characteristics from both the TRMM PR and GR observations can be obtained. The fused reflectivity data are evaluated based on a convective precipitation event through comparison with the high resolution TRMM PR and GR data with an interpolation algorithm.展开更多
Action recognition is important for understanding the human behaviors in the video,and the video representation is the basis for action recognition.This paper provides a new video representation based on convolution n...Action recognition is important for understanding the human behaviors in the video,and the video representation is the basis for action recognition.This paper provides a new video representation based on convolution neural networks（CNN）.For capturing human motion information in one CNN,we take both the optical flow maps and gray images as input,and combine multiple convolutional features by max pooling across frames.In another CNN,we input single color frame to capture context information.Finally,we take the top full connected layer vectors as video representation and train the classifiers by linear support vector machine.The experimental results show that the representation which integrates the optical flow maps and gray images obtains more discriminative properties than those which depend on only one element.On the most challenging data sets HMDB51 and UCF101,this video representation obtains competitive performance.展开更多
Based on the CNN-LSTM fusion deep neural network,this paper proposes a seismic velocity model building method that can simultaneously estimate the root mean square(RMS)velocity and interval velocity from the common-mi...Based on the CNN-LSTM fusion deep neural network,this paper proposes a seismic velocity model building method that can simultaneously estimate the root mean square(RMS)velocity and interval velocity from the common-midpoint(CMP)gather.In the proposed method,a convolutional neural network(CNN)Encoder and two long short-term memory networks(LSTMs)are used to extract spatial and temporal features from seismic signals,respectively,and a CNN Decoder is used to recover RMS velocity and interval velocity of underground media from various feature vectors.To address the problems of unstable gradients and easily fall into a local minimum in the deep neural network training process,we propose to use Kaiming normal initialization with zero negative slopes of rectifi ed units and to adjust the network learning process by optimizing the mean square error(MSE)loss function with the introduction of a freezing factor.The experiments on testing dataset show that CNN-LSTM fusion deep neural network can predict RMS velocity as well as interval velocity more accurately,and its inversion accuracy is superior to that of single neural network models.The predictions on the complex structures and Marmousi model are consistent with the true velocity variation trends,and the predictions on fi eld data can eff ectively correct the phase axis,improve the lateral continuity of phase axis and quality of stack section,indicating the eff ectiveness and decent generalization capability of the proposed method.展开更多
In wireless sensor networks, target classification differs from that in centralized sensing systems because of the distributed detection, wireless communication and limited resources. We study the classification probl...In wireless sensor networks, target classification differs from that in centralized sensing systems because of the distributed detection, wireless communication and limited resources. We study the classification problem of moving vehicles in wireless sensor networks using acoustic signals emitted from vehicles. Three algorithms including wavelet decomposition, weighted k-nearest-neighbor and Dempster-Shafer theory are combined in this paper. Finally, we use real world experimental data to validate the classification methods. The result shows that wavelet based feature extraction method can extract stable features from acoustic signals. By fusion with Dempster's rule, the classification performance is improved.展开更多
In practical multi-sensor information fusion systems, there exists uncertainty about the network structure, active state of sensors, and information itself （including fuzziness, randomness, incompleteness as well as ...In practical multi-sensor information fusion systems, there exists uncertainty about the network structure, active state of sensors, and information itself （including fuzziness, randomness, incompleteness as well as roughness, etc）. Hence it requires investigating the problem of uncertain information fusion. Robust learning algorithm which adapts to complex environment and the fuzzy inference algorithm which disposes fuzzy information are explored to solve the problem. Based on the fusion technology of neural networks and fuzzy inference algorithm, a multi-sensor uncertain information fusion system is modeled. Also RANFIS learning algorithm and fusing weight synthesized inference algorithm are developed from the ANFIS algorithm according to the concept of robust neural networks. This fusion system mainly consists of RANFIS confidence estimator, fusing weight synthesized inference knowledge base and weighted fusion section. The simulation result demonstrates that the proposed fusion model and algorithm have the capability of uncertain information fusion, thus is obviously advantageous compared with the conventional Kalman weighted fusion algorithm.展开更多
Scene recognition is a popular open problem in the computer vision field.Among lots of methods proposed in recent years,Convolutional Neural Network(CNN)based approaches achieve the best performance in scene recogniti...Scene recognition is a popular open problem in the computer vision field.Among lots of methods proposed in recent years,Convolutional Neural Network(CNN)based approaches achieve the best performance in scene recognition.We propose in this paper an advanced feature fusion algorithm using Multiple Convolutional Neural Network(Multi-CNN)for scene recognition.Unlike existing works that usually use individual convolutional neural network,a fusion of multiple different convolutional neural networks is applied for scene recognition.Firstly,we split training images in two directions and apply to three deep CNN model,and then extract features from the last full-connected(FC)layer and probabilistic layer on each model.Finally,feature vectors are fused with different fusion strategies in groups forwarded into SoftMax classifier.Our proposed algorithm is evaluated on three scene datasets for scene recognition.The experimental results demonstrate the effectiveness of proposed algorithm compared with other state-of-art approaches.展开更多
Multiple complex networks, each with different properties and mutually fused, have the problems that the evolving process is time varying and non-equilibrium, network structures are layered and interlacing, and evolvi...Multiple complex networks, each with different properties and mutually fused, have the problems that the evolving process is time varying and non-equilibrium, network structures are layered and interlacing, and evolving characteristics are difficult to be measured. On that account, a dynamic evolving model of complex network with fusion nodes and overlap edges（CNFNOEs） is proposed. Firstly, we define some related concepts of CNFNOEs, and analyze the conversion process of fusion relationship and hierarchy relationship. According to the property difference of various nodes and edges, fusion nodes and overlap edges are subsequently split, and then the CNFNOEs is transformed to interlacing layered complex networks（ILCN）. Secondly,the node degree saturation and attraction factors are defined. On that basis, the evolution algorithm and the local world evolution model for ILCN are put forward. Moreover, four typical situations of nodes evolution are discussed, and the degree distribution law during evolution is analyzed by means of the mean field method.Numerical simulation results show that nodes unreached degree saturation follow the exponential distribution with an error of no more than 6%; nodes reached degree saturation follow the distribution of their connection capacities with an error of no more than 3%; network weaving coefficients have a positive correlation with the highest probability of new node and initial number of connected edges. The results have verified the feasibility and effectiveness of the model, which provides a new idea and method for exploring CNFNOE＇s evolving process and law. Also, the model has good application prospects in structure and dynamics research of transportation network, communication network, social contact network,etc.展开更多
Obtaining comprehensive and accurate information is very important in intelligent tragic system （ITS）. In ITS, the GPS floating car system is an important approach for traffic data acquisition. However, in this syst...Obtaining comprehensive and accurate information is very important in intelligent tragic system （ITS）. In ITS, the GPS floating car system is an important approach for traffic data acquisition. However, in this system, the GPS blind areas caused by tall buildings or tunnels could affect the acquisition of tragic information and depress the system performance. Aiming at this problem, a novel method employing a back propagation （BP） neural network is developed to estimate the traffic speed in the GPS blind areas. When the speed of one road section is lost, the speed of its related road sections can be used to estimate its speed. The complete historical data of these road sections are used to train the neural network, using Levenberg-Marquardt learning algorithm. Then, the current speed of the related roads is used by the trained neural network to get the speed of the road section without GPS signal. We compare the speed of the road section estimated by our method with the real speed of this road section, and the experimental results show that the proposed method of traffic speed estimation is very effective.展开更多
In order to improve the detection accuracy of small objects,a neighborhood fusion-based hierarchical parallel feature pyramid network(NFPN)is proposed.Unlike the layer-by-layer structure adopted in the feature pyramid...In order to improve the detection accuracy of small objects,a neighborhood fusion-based hierarchical parallel feature pyramid network(NFPN)is proposed.Unlike the layer-by-layer structure adopted in the feature pyramid network(FPN)and deconvolutional single shot detector(DSSD),where the bottom layer of the feature pyramid network relies on the top layer,NFPN builds the feature pyramid network with no connections between the upper and lower layers.That is,it only fuses shallow features on similar scales.NFPN is highly portable and can be embedded in many models to further boost performance.Extensive experiments on PASCAL VOC 2007,2012,and COCO datasets demonstrate that the NFPN-based SSD without intricate tricks can exceed the DSSD model in terms of detection accuracy and inference speed,especially for small objects,e.g.,4%to 5%higher mAP(mean average precision)than SSD,and 2%to 3%higher mAP than DSSD.On VOC 2007 test set,the NFPN-based SSD with 300×300 input reaches 79.4%mAP at 34.6 frame/s,and the mAP can raise to 82.9%after using the multi-scale testing strategy.展开更多
基金supported by the National Natural Science Foundation of China(62276092,62303167)the Postdoctoral Fellowship Program(Grade C)of China Postdoctoral Science Foundation(GZC20230707)+3 种基金the Key Science and Technology Program of Henan Province,China(242102211051,242102211042,212102310084)Key Scientiffc Research Projects of Colleges and Universities in Henan Province,China(25A520009)the China Postdoctoral Science Foundation(2024M760808)the Henan Province medical science and technology research plan joint construction project(LHGJ2024069).
文摘Feature fusion is an important technique in medical image classification that can improve diagnostic accuracy by integrating complementary information from multiple sources.Recently,Deep Learning(DL)has been widely used in pulmonary disease diagnosis,such as pneumonia and tuberculosis.However,traditional feature fusion methods often suffer from feature disparity,information loss,redundancy,and increased complexity,hindering the further extension of DL algorithms.To solve this problem,we propose a Graph-Convolution Fusion Network with Self-Supervised Feature Alignment(Self-FAGCFN)to address the limitations of traditional feature fusion methods in deep learning-based medical image classification for respiratory diseases such as pneumonia and tuberculosis.The network integrates Convolutional Neural Networks(CNNs)for robust feature extraction from two-dimensional grid structures and Graph Convolutional Networks(GCNs)within a Graph Neural Network branch to capture features based on graph structure,focusing on significant node representations.Additionally,an Attention-Embedding Ensemble Block is included to capture critical features from GCN outputs.To ensure effective feature alignment between pre-and post-fusion stages,we introduce a feature alignment loss that minimizes disparities.Moreover,to address the limitations of proposed methods,such as inappropriate centroid discrepancies during feature alignment and class imbalance in the dataset,we develop a Feature-Centroid Fusion(FCF)strategy and a Multi-Level Feature-Centroid Update(MLFCU)algorithm,respectively.Extensive experiments on public datasets LungVision and Chest-Xray demonstrate that the Self-FAGCFN model significantly outperforms existing methods in diagnosing pneumonia and tuberculosis,highlighting its potential for practical medical applications.
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 the National Natural Science Foundation of China under Grant Nos.U21A20464,62066005Innovation Project of Guangxi Graduate Education under Grant No.YCSW2024313.
文摘Wireless sensor network deployment optimization is a classic NP-hard problem and a popular topic in academic research.However,the current research on wireless sensor network deployment problems uses overly simplistic models,and there is a significant gap between the research results and actual wireless sensor networks.Some scholars have now modeled data fusion networks to make them more suitable for practical applications.This paper will explore the deployment problem of a stochastic data fusion wireless sensor network(SDFWSN),a model that reflects the randomness of environmental monitoring and uses data fusion techniques widely used in actual sensor networks for information collection.The deployment problem of SDFWSN is modeled as a multi-objective optimization problem.The network life cycle,spatiotemporal coverage,detection rate,and false alarm rate of SDFWSN are used as optimization objectives to optimize the deployment of network nodes.This paper proposes an enhanced multi-objective mongoose optimization algorithm(EMODMOA)to solve the deployment problem of SDFWSN.First,to overcome the shortcomings of the DMOA algorithm,such as its low convergence and tendency to get stuck in a local optimum,an encircling and hunting strategy is introduced into the original algorithm to propose the EDMOA algorithm.The EDMOA algorithm is designed as the EMODMOA algorithm by selecting reference points using the K-Nearest Neighbor(KNN)algorithm.To verify the effectiveness of the proposed algorithm,the EMODMOA algorithm was tested at CEC 2020 and achieved good results.In the SDFWSN deployment problem,the algorithm was compared with the Non-dominated Sorting Genetic Algorithm II(NSGAII),Multiple Objective Particle Swarm Optimization(MOPSO),Multi-Objective Evolutionary Algorithm based on Decomposition(MOEA/D),and Multi-Objective Grey Wolf Optimizer(MOGWO).By comparing and analyzing the performance evaluation metrics and optimization results of the objective functions of the multi-objective algorithms,the algorithm outperforms the other algorithms in the SDFWSN deployment results.To better demonstrate the superiority of the algorithm,simulations of diverse test cases were also performed,and good results were obtained.
基金supported by National Natural Science Foundation of China(Nos.12105080 and 12375123)China Postdoctoral Science Foundation(No.2023M731015)Natural Science Foundation of Henan Province(No.242300422048).
文摘A machine learning approach based on Bayesian neural networks was developed to predict the complete fusion cross-sections of weakly bound nuclei.This method was trained and validated using 475 experimental data points from 39 reaction systems induced by ^(6,7)Li,^(9)Be,and ^(10)B.The constructed Bayesian neural network demonstrated a high degree of accuracy in evaluating complete fusion cross-sections.By comparing the predicted cross-sections with those obtained from a single-barrier penetration model,the suppression effect of ^(6,7)Li and ^(9)Be with a stable nucleus was systematically analyzed.In the cases of ^(6)Li and ^(7)Li,less suppression was predicted for relatively light-mass targets than for heavy-mass targets,and a notably distinct dependence relationship was identified,suggesting that the predominant breakup mechanisms might change in different mass target regions.In addition,minimum suppression factors were predicted to occur near target nuclei with neutron-closed shell.
基金supported by Qingdao Huanghai University School-Level ScientificResearch Project(2023KJ14)Undergraduate Teaching Reform Research Project of Shandong Provincial Department of Education(M2022328)+1 种基金National Natural Science Foundation of China under Grant(42472324)Qingdao Postdoctoral Foundation under Grant(QDBSH202402049).
文摘Multimodal image fusion plays an important role in image analysis and applications.Multimodal medical image fusion helps to combine contrast features from two or more input imaging modalities to represent fused information in a single image.One of the critical clinical applications of medical image fusion is to fuse anatomical and functional modalities for rapid diagnosis of malignant tissues.This paper proposes a multimodal medical image fusion network(MMIF-Net)based on multiscale hybrid attention.The method first decomposes the original image to obtain the low-rank and significant parts.Then,to utilize the features at different scales,we add amultiscalemechanism that uses three filters of different sizes to extract the features in the encoded network.Also,a hybrid attention module is introduced to obtain more image details.Finally,the fused images are reconstructed by decoding the network.We conducted experiments with clinical images from brain computed tomography/magnetic resonance.The experimental results show that the multimodal medical image fusion network method based on multiscale hybrid attention works better than other advanced fusion methods.
基金supported by Chongqing University of Posts and Telecommunications Ph.D.Innovative Talents Project(Grant No.BYJS202106)Chongqing Postgraduate Research Innovation Project(Grant No.CYB21203).
文摘Speech Emotion Recognition(SER)has received widespread attention as a crucial way for understanding human emotional states.However,the impact of irrelevant information on speech signals and data sparsity limit the development of SER system.To address these issues,this paper proposes a framework that incorporates the Attentive Mask Residual Network(AM-ResNet)and the self-supervised learning model Wav2vec 2.0 to obtain AM-ResNet features and Wav2vec 2.0 features respectively,together with a cross-attention module to interact and fuse these two features.The AM-ResNet branch mainly consists of maximum amplitude difference detection,mask residual block,and an attention mechanism.Among them,the maximum amplitude difference detection and the mask residual block act on the pre-processing and the network,respectively,to reduce the impact of silent frames,and the attention mechanism assigns different weights to unvoiced and voiced speech to reduce redundant emotional information caused by unvoiced speech.In the Wav2vec 2.0 branch,this model is introduced as a feature extractor to obtain general speech features(Wav2vec 2.0 features)through pre-training with a large amount of unlabeled speech data,which can assist the SER task and cope with data sparsity problems.In the cross-attention module,AM-ResNet features and Wav2vec 2.0 features are interacted with and fused to obtain the cross-fused features,which are used to predict the final emotion.Furthermore,multi-label learning is also used to add ambiguous emotion utterances to deal with data limitations.Finally,experimental results illustrate the usefulness and superiority of our proposed framework over existing state-of-the-art approaches.
文摘Skin cancer is the most prevalent cancer globally,primarily due to extensive exposure to Ultraviolet(UV)radiation.Early identification of skin cancer enhances the likelihood of effective treatment,as delays may lead to severe tumor advancement.This study proposes a novel hybrid deep learning strategy to address the complex issue of skin cancer diagnosis,with an architecture that integrates a Vision Transformer,a bespoke convolutional neural network(CNN),and an Xception module.They were evaluated using two benchmark datasets,HAM10000 and Skin Cancer ISIC.On the HAM10000,the model achieves a precision of 95.46%,an accuracy of 96.74%,a recall of 96.27%,specificity of 96.00%and an F1-Score of 95.86%.It obtains an accuracy of 93.19%,a precision of 93.25%,a recall of 92.80%,a specificity of 92.89%and an F1-Score of 93.19%on the Skin Cancer ISIC dataset.The findings demonstrate that the model that was proposed is robust and trustworthy when it comes to the classification of skin lesions.In addition,the utilization of Explainable AI techniques,such as Grad-CAM visualizations,assists in highlighting the most significant lesion areas that have an impact on the decisions that are made by the model.
基金funded by the Research on SOC/SOH Joint Estimation Technology of Electric Vehicle Battery System State Based on Online Parameter Identification Project(2019)the National Natural Science Foundation of China(Grant No.51877120)。
文摘The accurate state of health(SOH)estimation of lithium-ion batteries is crucial for efficient,healthy,and safe operation of battery systems.Extracting meaningful aging information from highly stochastic and noisy data segments while designing SOH estimation algorithms that efficiently handle the large-scale computational demands of cloud-based battery management systems presents a substantial challenge.In this work,we propose a quantum convolutional neural network(QCNN)model designed for accurate,robust,and generalizable SOH estimation with minimal data and parameter requirements and is compatible with quantum computing cloud platforms in the Noisy Intermediate-Scale Quantum.First,we utilize data from 4 datasets comprising 272 cells,covering 5 chemical compositions,4 rated parameters,and 73operating conditions.We design 5 voltage windows as small as 0.3 V for each cell from incremental capacity peaks for stochastic SOH estimation scenarios generation.We extract 3 effective health indicators(HIs)sequences and develop an automated feature fusion method using quantum rotation gate encoding,achieving an R2of 96%.Subsequently,we design a QCNN whose convolutional layer,constructed with variational quantum circuits,comprises merely 39 parameters.Additionally,we explore the impact of training set size,using strategies,and battery materials on the model’s accuracy.Finally,the QCNN with quantum convolutional layers reduces root mean squared error by 28% and achieves an R^(2)exceeding 96% compared to other three commonly used algorithms.This work demonstrates the effectiveness of quantum encoding for automated feature fusion of HIs extracted from limited discharge data.It highlights the potential of QCNN in improving the accuracy,robustness,and generalization of SOH estimation while dealing with stochastic and noisy data with few parameters and simple structure.It also suggests a new paradigm for leveraging quantum computational power in SOH estimation.
基金Supported by Sichuan Science and Technology Program(2023YFSY0026,2023YFH0004).
文摘3D medical image reconstruction has significantly enhanced diagnostic accuracy,yet the reliance on densely sampled projection data remains a major limitation in clinical practice.Sparse-angle X-ray imaging,though safer and faster,poses challenges for accurate volumetric reconstruction due to limited spatial information.This study proposes a 3D reconstruction neural network based on adaptive weight fusion(AdapFusionNet)to achieve high-quality 3D medical image reconstruction from sparse-angle X-ray images.To address the issue of spatial inconsistency in multi-angle image reconstruction,an innovative adaptive fusion module was designed to score initial reconstruction results during the inference stage and perform weighted fusion,thereby improving the final reconstruction quality.The reconstruction network is built on an autoencoder(AE)framework and uses orthogonal-angle X-ray images(frontal and lateral projections)as inputs.The encoder extracts 2D features,which the decoder maps into 3D space.This study utilizes a lung CT dataset to obtain complete three-dimensional volumetric data,from which digitally reconstructed radiographs(DRR)are generated at various angles to simulate X-ray images.Since real-world clinical X-ray images rarely come with perfectly corresponding 3D“ground truth,”using CT scans as the three-dimensional reference effectively supports the training and evaluation of deep networks for sparse-angle X-ray 3D reconstruction.Experiments conducted on the LIDC-IDRI dataset with simulated X-ray images(DRR images)as training data demonstrate the superior performance of AdapFusionNet compared to other fusion methods.Quantitative results show that AdapFusionNet achieves SSIM,PSNR,and MAE values of 0.332,13.404,and 0.163,respectively,outperforming other methods(SingleViewNet:0.289,12.363,0.182;AvgFusionNet:0.306,13.384,0.159).Qualitative analysis further confirms that AdapFusionNet significantly enhances the reconstruction of lung and chest contours while effectively reducing noise during the reconstruction process.The findings demonstrate that AdapFusionNet offers significant advantages in 3D reconstruction of sparse-angle X-ray images.
基金supported by the National Natural Science Foundation of China(Grant Nos.12272259 and 52005148).
文摘An intelligent diagnosis method based on self-adaptiveWasserstein dual generative adversarial networks and feature fusion is proposed due to problems such as insufficient sample size and incomplete fault feature extraction,which are commonly faced by rolling bearings and lead to low diagnostic accuracy.Initially,dual models of the Wasserstein deep convolutional generative adversarial network incorporating gradient penalty(1D-2DWDCGAN)are constructed to augment the original dataset.A self-adaptive loss threshold control training strategy is introduced,and establishing a self-adaptive balancing mechanism for stable model training.Subsequently,a diagnostic model based on multidimensional feature fusion is designed,wherein complex features from various dimensions are extracted,merging the original signal waveform features,structured features,and time-frequency features into a deep composite feature representation that encompasses multiple dimensions and scales;thus,efficient and accurate small sample fault diagnosis is facilitated.Finally,an experiment between the bearing fault dataset of CaseWestern ReserveUniversity and the fault simulation experimental platformdataset of this research group shows that this method effectively supplements the dataset and remarkably improves the diagnostic accuracy.The diagnostic accuracy after data augmentation reached 99.94%and 99.87%in two different experimental environments,respectively.In addition,robustness analysis is conducted on the diagnostic accuracy of the proposed method under different noise backgrounds,verifying its good generalization performance.
基金The National Natural Science Foundation of China(No.60875035)the Natural Science Foundation of Jiangsu Province(No.BK2008294)the National High Technology Research and Development Program of China(863 Program)(No.2006AA05A107)
文摘A constructive-pruning hybrid method （CPHM） for radial basis function （RBF） networks is proposed to improve the prediction accuracy of ash fusion temperatures （AFT）. The CPHM incorporates the advantages of the construction algorithm and the pruning algorithm of neural networks, and the training process of the CPHM is divided into two stages： rough tuning and fine tuning. In rough tuning, new hidden units are added to the current network until some performance index is satisfied. In fine tuning, the network structure and the model parameters are further adjusted. And, based on components of coal ash, a model using the CPHM is established to predict the AFT. The results show that the CPHM prediction model is characterized by its high precision, compact network structure, as well as strong generalization ability and robustness.
基金supported by funding from the Natural Science Foundation of Jiangsu Province (Grant No. BK20171457)the 2013 Special Fund for Meteorological Scientific Research in the Public Interest (Grant No. GYHY201306078)+1 种基金the National Natural Science Foundation of China (Grant No. 41301399)Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD)
文摘The spaceborne precipitation radar onboard the Tropical Rainfall Measuring Mission satellite （TRMM PR） can provide good measurement of the vertical structure of reflectivity, while ground radar （GR） has a relatively high horizontal resolution and greater sensitivity. Fusion of TRMM PR and GR reflectivity data may maximize the advantages from both instruments. In this paper, TRMM PR and GR reflectivity data are fused using a neural network （NN）-based approach. The main steps included are： quality control of TRMM PR and GR reflectivity data; spatiotemporal matchup; GR calibration bias correction; conversion of TRMM PR data from Ku to S band; fusion of TRMM PR and GR reflectivity data with an NN method： interpolation of reflectivity data that are below PR＇s sensitivity; blind areas compensation with a distance weighting-based merging approach; combination of three types of data： data with the NN method, data below PR＇s sensitivity and data within compensated blind areas. During the NN fusion step, the TRMM PR data are taken as targets of the training NNs, and gridded GR data after horizontal downsampling at different heights are used as the input. The trained NNs are then used to obtain 3D high-resolution reflectivity from the original GR gridded data. After 3D fusion of the TRMM PR and GR reflectivity data, a more complete and finer-scale 3D radar reflectivity dataset incorporating characteristics from both the TRMM PR and GR observations can be obtained. The fused reflectivity data are evaluated based on a convective precipitation event through comparison with the high resolution TRMM PR and GR data with an interpolation algorithm.
基金Supported by the National High Technology Research and Development Program of China(863 Program,2015AA016306)National Nature Science Foundation of China(61231015)+2 种基金Internet of Things Development Funding Project of Ministry of Industry in 2013(25)Technology Research Program of Ministry of Public Security(2016JSYJA12)the Nature Science Foundation of Hubei Province(2014CFB712)
文摘Action recognition is important for understanding the human behaviors in the video,and the video representation is the basis for action recognition.This paper provides a new video representation based on convolution neural networks（CNN）.For capturing human motion information in one CNN,we take both the optical flow maps and gray images as input,and combine multiple convolutional features by max pooling across frames.In another CNN,we input single color frame to capture context information.Finally,we take the top full connected layer vectors as video representation and train the classifiers by linear support vector machine.The experimental results show that the representation which integrates the optical flow maps and gray images obtains more discriminative properties than those which depend on only one element.On the most challenging data sets HMDB51 and UCF101,this video representation obtains competitive performance.
基金financially supported by the Key Project of National Natural Science Foundation of China (No. 41930431)the Project of National Natural Science Foundation of China (Nos. 41904121, 41804133, and 41974116)Joint Guidance Project of Natural Science Foundation of Heilongjiang Province (No. LH2020D006)
文摘Based on the CNN-LSTM fusion deep neural network,this paper proposes a seismic velocity model building method that can simultaneously estimate the root mean square(RMS)velocity and interval velocity from the common-midpoint(CMP)gather.In the proposed method,a convolutional neural network(CNN)Encoder and two long short-term memory networks(LSTMs)are used to extract spatial and temporal features from seismic signals,respectively,and a CNN Decoder is used to recover RMS velocity and interval velocity of underground media from various feature vectors.To address the problems of unstable gradients and easily fall into a local minimum in the deep neural network training process,we propose to use Kaiming normal initialization with zero negative slopes of rectifi ed units and to adjust the network learning process by optimizing the mean square error(MSE)loss function with the introduction of a freezing factor.The experiments on testing dataset show that CNN-LSTM fusion deep neural network can predict RMS velocity as well as interval velocity more accurately,and its inversion accuracy is superior to that of single neural network models.The predictions on the complex structures and Marmousi model are consistent with the true velocity variation trends,and the predictions on fi eld data can eff ectively correct the phase axis,improve the lateral continuity of phase axis and quality of stack section,indicating the eff ectiveness and decent generalization capability of the proposed method.
基金Supported in part by Science & Technology Department of Shanghai (05dz15004)
文摘In wireless sensor networks, target classification differs from that in centralized sensing systems because of the distributed detection, wireless communication and limited resources. We study the classification problem of moving vehicles in wireless sensor networks using acoustic signals emitted from vehicles. Three algorithms including wavelet decomposition, weighted k-nearest-neighbor and Dempster-Shafer theory are combined in this paper. Finally, we use real world experimental data to validate the classification methods. The result shows that wavelet based feature extraction method can extract stable features from acoustic signals. By fusion with Dempster's rule, the classification performance is improved.
基金This project was supported by the National Natural Science Foundation of China (60572038)
文摘In practical multi-sensor information fusion systems, there exists uncertainty about the network structure, active state of sensors, and information itself （including fuzziness, randomness, incompleteness as well as roughness, etc）. Hence it requires investigating the problem of uncertain information fusion. Robust learning algorithm which adapts to complex environment and the fuzzy inference algorithm which disposes fuzzy information are explored to solve the problem. Based on the fusion technology of neural networks and fuzzy inference algorithm, a multi-sensor uncertain information fusion system is modeled. Also RANFIS learning algorithm and fusing weight synthesized inference algorithm are developed from the ANFIS algorithm according to the concept of robust neural networks. This fusion system mainly consists of RANFIS confidence estimator, fusing weight synthesized inference knowledge base and weighted fusion section. The simulation result demonstrates that the proposed fusion model and algorithm have the capability of uncertain information fusion, thus is obviously advantageous compared with the conventional Kalman weighted fusion algorithm.
文摘Scene recognition is a popular open problem in the computer vision field.Among lots of methods proposed in recent years,Convolutional Neural Network(CNN)based approaches achieve the best performance in scene recognition.We propose in this paper an advanced feature fusion algorithm using Multiple Convolutional Neural Network(Multi-CNN)for scene recognition.Unlike existing works that usually use individual convolutional neural network,a fusion of multiple different convolutional neural networks is applied for scene recognition.Firstly,we split training images in two directions and apply to three deep CNN model,and then extract features from the last full-connected(FC)layer and probabilistic layer on each model.Finally,feature vectors are fused with different fusion strategies in groups forwarded into SoftMax classifier.Our proposed algorithm is evaluated on three scene datasets for scene recognition.The experimental results demonstrate the effectiveness of proposed algorithm compared with other state-of-art approaches.
基金supported by the National Natural Science Foundation of China(615730176140149961174162)
文摘Multiple complex networks, each with different properties and mutually fused, have the problems that the evolving process is time varying and non-equilibrium, network structures are layered and interlacing, and evolving characteristics are difficult to be measured. On that account, a dynamic evolving model of complex network with fusion nodes and overlap edges（CNFNOEs） is proposed. Firstly, we define some related concepts of CNFNOEs, and analyze the conversion process of fusion relationship and hierarchy relationship. According to the property difference of various nodes and edges, fusion nodes and overlap edges are subsequently split, and then the CNFNOEs is transformed to interlacing layered complex networks（ILCN）. Secondly,the node degree saturation and attraction factors are defined. On that basis, the evolution algorithm and the local world evolution model for ILCN are put forward. Moreover, four typical situations of nodes evolution are discussed, and the degree distribution law during evolution is analyzed by means of the mean field method.Numerical simulation results show that nodes unreached degree saturation follow the exponential distribution with an error of no more than 6%; nodes reached degree saturation follow the distribution of their connection capacities with an error of no more than 3%; network weaving coefficients have a positive correlation with the highest probability of new node and initial number of connected edges. The results have verified the feasibility and effectiveness of the model, which provides a new idea and method for exploring CNFNOE＇s evolving process and law. Also, the model has good application prospects in structure and dynamics research of transportation network, communication network, social contact network,etc.
基金funded by National Key Technology R&D Program of China (No.2006BAG01A03)
文摘Obtaining comprehensive and accurate information is very important in intelligent tragic system （ITS）. In ITS, the GPS floating car system is an important approach for traffic data acquisition. However, in this system, the GPS blind areas caused by tall buildings or tunnels could affect the acquisition of tragic information and depress the system performance. Aiming at this problem, a novel method employing a back propagation （BP） neural network is developed to estimate the traffic speed in the GPS blind areas. When the speed of one road section is lost, the speed of its related road sections can be used to estimate its speed. The complete historical data of these road sections are used to train the neural network, using Levenberg-Marquardt learning algorithm. Then, the current speed of the related roads is used by the trained neural network to get the speed of the road section without GPS signal. We compare the speed of the road section estimated by our method with the real speed of this road section, and the experimental results show that the proposed method of traffic speed estimation is very effective.
基金The National Natural Science Foundation of China(No.61603091)。
文摘In order to improve the detection accuracy of small objects,a neighborhood fusion-based hierarchical parallel feature pyramid network(NFPN)is proposed.Unlike the layer-by-layer structure adopted in the feature pyramid network(FPN)and deconvolutional single shot detector(DSSD),where the bottom layer of the feature pyramid network relies on the top layer,NFPN builds the feature pyramid network with no connections between the upper and lower layers.That is,it only fuses shallow features on similar scales.NFPN is highly portable and can be embedded in many models to further boost performance.Extensive experiments on PASCAL VOC 2007,2012,and COCO datasets demonstrate that the NFPN-based SSD without intricate tricks can exceed the DSSD model in terms of detection accuracy and inference speed,especially for small objects,e.g.,4%to 5%higher mAP(mean average precision)than SSD,and 2%to 3%higher mAP than DSSD.On VOC 2007 test set,the NFPN-based SSD with 300×300 input reaches 79.4%mAP at 34.6 frame/s,and the mAP can raise to 82.9%after using the multi-scale testing strategy.
