With the rapid advancement of social economies,intelligent transportation systems are gaining increasing atten-tion.Central to these systems is the detection of abnormal vehicle behavior,which remains a critical chall...With the rapid advancement of social economies,intelligent transportation systems are gaining increasing atten-tion.Central to these systems is the detection of abnormal vehicle behavior,which remains a critical challenge due to the complexity of urban roadways and the variability of external conditions.Current research on detecting abnormal traffic behaviors is still nascent,with significant room for improvement in recognition accuracy.To address this,this research has developed a new model for recognizing abnormal traffic behaviors.This model employs the R3D network as its core architecture,incorporating a dense block to facilitate feature reuse.This approach not only enhances performance with fewer parameters and reduced computational demands but also allows for the acquisition of new features while simplifying the overall network structure.Additionally,this research integrates a self-attentive method that dynamically adjusts to the prevailing traffic conditions,optimizing the relevance of features for the task at hand.For temporal analysis,a Bi-LSTM layer is utilized to extract and learn from time-based data nuances.This research conducted a series of comparative experiments using the UCF-Crime dataset,achieving a notable accuracy of 89.30%on our test set.Our results demonstrate that our model not only operates with fewer parameters but also achieves superior recognition accuracy compared to previous models.展开更多
The main task of magnetic resonance imaging (MRI) automatic brain tumor segmentation is to automaticallysegment the brain tumor edema, peritumoral edema, endoscopic core, enhancing tumor core and nonenhancingtumor cor...The main task of magnetic resonance imaging (MRI) automatic brain tumor segmentation is to automaticallysegment the brain tumor edema, peritumoral edema, endoscopic core, enhancing tumor core and nonenhancingtumor core from 3D MR images. Because the location, size, shape and intensity of brain tumors vary greatly, itis very difficult to segment these brain tumor regions automatically. In this paper, by combining the advantagesof DenseNet and ResNet, we proposed a new 3D U-Net with dense encoder blocks and residual decoder blocks.We used dense blocks in the encoder part and residual blocks in the decoder part. The number of output featuremaps increases with the network layers in contracting path of encoder, which is consistent with the characteristicsof dense blocks. Using dense blocks can decrease the number of network parameters, deepen network layers,strengthen feature propagation, alleviate vanishing-gradient and enlarge receptive fields. The residual blockswere used in the decoder to replace the convolution neural block of original U-Net, which made the networkperformance better. Our proposed approach was trained and validated on the BraTS2019 training and validationdata set. We obtained dice scores of 0.901, 0.815 and 0.766 for whole tumor, tumor core and enhancing tumorcore respectively on the BraTS2019 validation data set. Our method has the better performance than the original3D U-Net. The results of our experiment demonstrate that compared with some state-of-the-art methods, ourapproach is a competitive automatic brain tumor segmentation method.展开更多
Generative adversarial networks(GANs)are paid more attention to dealing with the end-to-end speech enhancement in recent years.Various GANbased enhancement methods are presented to improve the quality of reconstructed...Generative adversarial networks(GANs)are paid more attention to dealing with the end-to-end speech enhancement in recent years.Various GANbased enhancement methods are presented to improve the quality of reconstructed speech.However,the performance of these GAN-based methods is worse than those of masking-based methods.To tackle this problem,we propose speech enhancement method with a residual dense generative adversarial network(RDGAN)contributing to map the log-power spectrum(LPS)of degraded speech to the clean one.In detail,a residual dense block(RDB)architecture is designed to better estimate the LPS of clean speech,which can extract rich local features of LPS through densely connected convolution layers.Meanwhile,sequential RDB connections are incorporated on various scales of LPS.It significantly increases the feature learning flexibility and robustness in the time-frequency domain.Simulations show that the proposed method achieves attractive speech enhancement performance in various acoustic environments.Specifically,in the untrained acoustic test with limited priors,e.g.,unmatched signal-to-noise ratio(SNR)and unmatched noise category,RDGAN can still outperform the existing GAN-based methods and masking-based method in the measures of PESQ and other evaluation indexes.It indicates that our method is more generalized in untrained conditions.展开更多
Fall behavior is closely related to high mortality in the elderly,so fall detection becomes an important and urgent research area.However,the existing fall detection methods are difficult to be applied in daily life d...Fall behavior is closely related to high mortality in the elderly,so fall detection becomes an important and urgent research area.However,the existing fall detection methods are difficult to be applied in daily life due to a large amount of calculation and poor detection accuracy.To solve the above problems,this paper proposes a dense spatial-temporal graph convolutional network based on lightweight OpenPose.Lightweight OpenPose uses MobileNet as a feature extraction network,and the prediction layer uses bottleneck-asymmetric structure,thus reducing the amount of the network.The bottleneck-asymmetrical structure compresses the number of input channels of feature maps by 1×1 convolution and replaces the 7×7 convolution structure with the asymmetric structure of 1×7 convolution,7×1 convolution,and 7×7 convolution in parallel.The spatial-temporal graph convolutional network divides the multi-layer convolution into dense blocks,and the convolutional layers in each dense block are connected,thus improving the feature transitivity,enhancing the network’s ability to extract features,thus improving the detection accuracy.Two representative datasets,Multiple Cameras Fall dataset(MCF),and Nanyang Technological University Red Green Blue+Depth Action Recognition dataset(NTU RGB+D),are selected for our experiments,among which NTU RGB+D has two evaluation benchmarks.The results show that the proposed model is superior to the current fall detection models.The accuracy of this network on the MCF dataset is 96.3%,and the accuracies on the two evaluation benchmarks of the NTU RGB+D dataset are 85.6%and 93.5%,respectively.展开更多
Masking-based and spectrum mapping-based methods are the two main algorithms of speech enhancement with deep neural network(DNN).But the mapping-based methods only utilizes the phase of noisy speech,which limits the u...Masking-based and spectrum mapping-based methods are the two main algorithms of speech enhancement with deep neural network(DNN).But the mapping-based methods only utilizes the phase of noisy speech,which limits the upper bound of speech enhancement performance.Maskingbased methods need to accurately estimate the masking which is still the key problem.Combining the advantages of above two types of methods,this paper proposes the speech enhancement algorithm MM-RDN(maskingmapping residual dense network)based on masking-mapping(MM)and residual dense network(RDN).Using the logarithmic power spectrogram(LPS)of consecutive frames,MM estimates the ideal ratio masking(IRM)matrix of consecutive frames.RDN can make full use of feature maps of all layers.Meanwhile,using the global residual learning to combine the shallow features and deep features,RDN obtains the global dense features from the LPS,thereby improves estimated accuracy of the IRM matrix.Simulations show that the proposed method achieves attractive speech enhancement performance in various acoustic environments.Specifically,in the untrained acoustic test with limited priors,e.g.,unmatched signal-to-noise ratio(SNR)and unmatched noise category,MM-RDN can still outperform the existing convolutional recurrent network(CRN)method in themeasures of perceptual evaluation of speech quality(PESQ)and other evaluation indexes.It indicates that the proposed algorithm is more generalized in untrained conditions.展开更多
In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of ea...In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of each layer were enhanced into the global features by the proposed residual dense block,which ensured that the generated images retain more details.Secondly,a multi-scale structure was adopted to extract multi-scale semantic features of the original images.Finally,the features obtained from the dual channels were fused by an adaptive fusion module to further optimize the features.The discriminant network adopted the structure of the Markov discriminator.In addition,by constructing mean square error,structural similarity,and perceived color loss function,the generated image is consistent with the reference image in structure,color,and content.The experimental results showed that the enhanced underwater image deblurring effect of the proposed algorithm was good and the problem of underwater image color bias was effectively improved.In both subjective and objective evaluation indexes,the experimental results of the proposed algorithm are better than those of the comparison algorithm.展开更多
Visible light positioning (VLP) is an emerging candidate for indoor positioning, which carl simultaneously meet the requirements for accuracy, cost, coverage area, and security. However, intercell interference causc...Visible light positioning (VLP) is an emerging candidate for indoor positioning, which carl simultaneously meet the requirements for accuracy, cost, coverage area, and security. However, intercell interference causcd by light intensity superposition linfits the application of VLP. In this Letter, we propose a united block sequence mapping (UBSM)-based VLP that utilizes superposition to integrate the multidimensional information from dense small cells into 2D information. The experimental result shows that UBSM-based VLP can achieve an accuracy of 1.5 cm with a 0.4 m row spacing and 0.35 m column spacing of LED lights.展开更多
WALKING on the streets of Jingrong Town in Chengdu,capital of Sichuan Province,bustling groups of people gather around the blocks of businesses there.This dense entrepreneurial climate sharply contrasts with the scene...WALKING on the streets of Jingrong Town in Chengdu,capital of Sichuan Province,bustling groups of people gather around the blocks of businesses there.This dense entrepreneurial climate sharply contrasts with the scene one year ago.Ghost Town Revitalized More than 2,000 years ago,Li Bing,chief of Shu Prefecture(roughly modern-day Sichuan),and his son built a water conservancy project in Dujiangyan,a small city in Sichuan Province.展开更多
To overcome the computational burden of processing three-dimensional(3 D)medical scans and the lack of spatial information in two-dimensional(2 D)medical scans,a novel segmentation method was proposed that integrates ...To overcome the computational burden of processing three-dimensional(3 D)medical scans and the lack of spatial information in two-dimensional(2 D)medical scans,a novel segmentation method was proposed that integrates the segmentation results of three densely connected 2 D convolutional neural networks(2 D-CNNs).In order to combine the lowlevel features and high-level features,we added densely connected blocks in the network structure design so that the low-level features will not be missed as the network layer increases during the learning process.Further,in order to resolve the problems of the blurred boundary of the glioma edema area,we superimposed and fused the T2-weighted fluid-attenuated inversion recovery(FLAIR)modal image and the T2-weighted(T2)modal image to enhance the edema section.For the loss function of network training,we improved the cross-entropy loss function to effectively avoid network over-fitting.On the Multimodal Brain Tumor Image Segmentation Challenge(BraTS)datasets,our method achieves dice similarity coefficient values of 0.84,0.82,and 0.83 on the BraTS2018 training;0.82,0.85,and 0.83 on the BraTS2018 validation;and 0.81,0.78,and 0.83 on the BraTS2013 testing in terms of whole tumors,tumor cores,and enhancing cores,respectively.Experimental results showed that the proposed method achieved promising accuracy and fast processing,demonstrating good potential for clinical medicine.展开更多
Many networks are designed to stack a large number of residual blocks,deepen the network and improve network performance through short residual connec-tion,long residual connection,and dense connection.However,without...Many networks are designed to stack a large number of residual blocks,deepen the network and improve network performance through short residual connec-tion,long residual connection,and dense connection.However,without consider-ing different contributions of different depth features to the network,these de-signs have the problem of evaluating the importance of different depth features.To solve this problem,this paper proposes an adaptive densely residual net-work(ADRNet)for the single image super resolution.ADRN realizes the evalua-tion of distributions of different depth features and learns more representative features.An adaptive densely residual block(ADRB)was designed,combining 3 residual blocks(RB)and dense connection was added.It learned the attention score of each dense connection through adaptive dense connections,and the at-tention score reflected the importance of the features of each RB.To further en-hance the performance of ADRB,a multi-direction attention block(MDAB)was introduced to obtain multidirectional context information.Through comparative experiments,it is proved that theproposed ADRNet is superior to the existing methods.Through ablation experiments,it is proved that evaluating features of different depths helps to improve network performance.展开更多
滚动轴承是风电机组关键部件,其运行工况复杂,故障类型难以准确识别。针对传统深度神经网络在强噪声环境下特征学习能力不足的问题,提出一种基于稠密连接模块的改进深度残差收缩网络(Deep residual shrinkage network based on dense bl...滚动轴承是风电机组关键部件,其运行工况复杂,故障类型难以准确识别。针对传统深度神经网络在强噪声环境下特征学习能力不足的问题,提出一种基于稠密连接模块的改进深度残差收缩网络(Deep residual shrinkage network based on dense block,DB-DRSN),实现强噪声、不同负载工况下滚动轴承故障的高效诊断。首先,将添加不同等级噪声的振动信号间隔采样并矩阵化,构建二维灰度图作为输入样本。然后,基于Dense block构造稠密连接残差收缩模块层(Residual shrinkage block unit based on dense block,DB-RSBU),利用Bottleneck层替代残差收缩模块中的卷积隐层,并加入Concat连接,达到对浅层和深层特征的充分利用。在每次稠密连接后通过1×1卷积进行降维,利用注意力模块和软阈值对逐通道特征赋不同阈值并降噪。最后,输入样本经过卷积池化层和DB-RSBU层堆叠的网络得到分类结果。试验表明,DB-DRSN模型在CWRU与PU滚动轴承数据集上不同噪声等级下的平均诊断准确率分别达到99.80%和96.44%,相比其他模型有更高的准确率、更快的收敛速度和更强的抗干扰能力。引入稠密连接核心思想对网络结构的改进可为基于数据驱动的风电机组滚动轴承故障诊断方法提供新思路。展开更多
To address the problems of lack of high-frequency information and texture details and unstable training in superresolution generative adversarial net-works,this paper optimizes the generator and discriminator based on...To address the problems of lack of high-frequency information and texture details and unstable training in superresolution generative adversarial net-works,this paper optimizes the generator and discriminator based on the SRGAN model.First,the residual dense block is used as the basic structural unit of the gen-erator to improve the network’s feature extraction capability.Second,enhanced lightweight coordinate attention is incorporated to help the network more precisely concentrate on high-frequency location information,thereby allowing the gener-ator to produce more realistic image reconstruction results.Then,we propose a symmetric and efficient pyramidal segmentation attention discriminator network in which the attention mechanism is capable of derivingfiner-grained multiscale spatial information and creating long-term dependencies between multiscale chan-nel attentions,thus enhancing the discriminative ability of the network.Finally,a Charbonnier loss function and a gradient variance loss function with improved robustness are used to better realize the image’s texture structure and enhance the model’s stability.Thefindings from the experiments reveal that the reconstructed image quality enhances the average peak signal-to-noise ratio(PSNR)by 1.59 dB and the structural similarity index(SSIM)by 0.045 when compared to SRGAN on the three test sets.Compared with the state-of-the-art methods,the reconstructed images have a clearer texture structure,richer high-frequency details,and better visual effects.展开更多
The superresolution(SR)method based on generative adversarial networks(GANs)cannot adequately capture enough diversity from training data,resulting in misalignment between input low resolution(LR)images and output hig...The superresolution(SR)method based on generative adversarial networks(GANs)cannot adequately capture enough diversity from training data,resulting in misalignment between input low resolution(LR)images and output high resolution(HR)images.GAN training has difficulty converging.Based on this,an advanced GAN-based image SR reconstructionmethod is presented.First,the dense connection residual block and attention mechanism are integrated into the GAN generator to improve high-frequency feature extraction.Meanwhile,an added discriminator is added into the GAN discriminant network,which forms a dual discriminator to ensure that the process of training is stable.Second,the more robust Charbonnier loss is used instead of the mean square error(MSE)loss to compare similarities between the obtained image and actual image,and the total variation(TV)loss is employed to smooth the training results.Finally,the experimental results indicate that global structures can be better reconstructed using the method of this paper and texture details of images compared with other SOTA methods.The peak signal-to-noise ratio(PSNR)values by the method of this paper are improved by an average of 2.24 dB,and the structural similarity index measure(SSIM)values are improved by an average of 0.07.展开更多
基金supported by the National Natural Science Foundation of China(61971007&61571013).
文摘With the rapid advancement of social economies,intelligent transportation systems are gaining increasing atten-tion.Central to these systems is the detection of abnormal vehicle behavior,which remains a critical challenge due to the complexity of urban roadways and the variability of external conditions.Current research on detecting abnormal traffic behaviors is still nascent,with significant room for improvement in recognition accuracy.To address this,this research has developed a new model for recognizing abnormal traffic behaviors.This model employs the R3D network as its core architecture,incorporating a dense block to facilitate feature reuse.This approach not only enhances performance with fewer parameters and reduced computational demands but also allows for the acquisition of new features while simplifying the overall network structure.Additionally,this research integrates a self-attentive method that dynamically adjusts to the prevailing traffic conditions,optimizing the relevance of features for the task at hand.For temporal analysis,a Bi-LSTM layer is utilized to extract and learn from time-based data nuances.This research conducted a series of comparative experiments using the UCF-Crime dataset,achieving a notable accuracy of 89.30%on our test set.Our results demonstrate that our model not only operates with fewer parameters but also achieves superior recognition accuracy compared to previous models.
基金This was supported partially by Sichuan Science and Technology Program under Grants 2019YJ0356,21ZDYF2484,21GJHZ0061Scientific Research Foundation of Education Department of Sichuan Province under Grant 18ZB0117.
文摘The main task of magnetic resonance imaging (MRI) automatic brain tumor segmentation is to automaticallysegment the brain tumor edema, peritumoral edema, endoscopic core, enhancing tumor core and nonenhancingtumor core from 3D MR images. Because the location, size, shape and intensity of brain tumors vary greatly, itis very difficult to segment these brain tumor regions automatically. In this paper, by combining the advantagesof DenseNet and ResNet, we proposed a new 3D U-Net with dense encoder blocks and residual decoder blocks.We used dense blocks in the encoder part and residual blocks in the decoder part. The number of output featuremaps increases with the network layers in contracting path of encoder, which is consistent with the characteristicsof dense blocks. Using dense blocks can decrease the number of network parameters, deepen network layers,strengthen feature propagation, alleviate vanishing-gradient and enlarge receptive fields. The residual blockswere used in the decoder to replace the convolution neural block of original U-Net, which made the networkperformance better. Our proposed approach was trained and validated on the BraTS2019 training and validationdata set. We obtained dice scores of 0.901, 0.815 and 0.766 for whole tumor, tumor core and enhancing tumorcore respectively on the BraTS2019 validation data set. Our method has the better performance than the original3D U-Net. The results of our experiment demonstrate that compared with some state-of-the-art methods, ourapproach is a competitive automatic brain tumor segmentation method.
基金This work is supported by the National Key Research and Development Program of China under Grant 2020YFC2004003 and Grant 2020YFC2004002the National Nature Science Foundation of China(NSFC)under Grant No.61571106。
文摘Generative adversarial networks(GANs)are paid more attention to dealing with the end-to-end speech enhancement in recent years.Various GANbased enhancement methods are presented to improve the quality of reconstructed speech.However,the performance of these GAN-based methods is worse than those of masking-based methods.To tackle this problem,we propose speech enhancement method with a residual dense generative adversarial network(RDGAN)contributing to map the log-power spectrum(LPS)of degraded speech to the clean one.In detail,a residual dense block(RDB)architecture is designed to better estimate the LPS of clean speech,which can extract rich local features of LPS through densely connected convolution layers.Meanwhile,sequential RDB connections are incorporated on various scales of LPS.It significantly increases the feature learning flexibility and robustness in the time-frequency domain.Simulations show that the proposed method achieves attractive speech enhancement performance in various acoustic environments.Specifically,in the untrained acoustic test with limited priors,e.g.,unmatched signal-to-noise ratio(SNR)and unmatched noise category,RDGAN can still outperform the existing GAN-based methods and masking-based method in the measures of PESQ and other evaluation indexes.It indicates that our method is more generalized in untrained conditions.
基金supported,in part,by the National Nature Science Foundation of China under Grant Numbers 62272236,62376128in part,by the Natural Science Foundation of Jiangsu Province under Grant Numbers BK20201136,BK20191401.
文摘Fall behavior is closely related to high mortality in the elderly,so fall detection becomes an important and urgent research area.However,the existing fall detection methods are difficult to be applied in daily life due to a large amount of calculation and poor detection accuracy.To solve the above problems,this paper proposes a dense spatial-temporal graph convolutional network based on lightweight OpenPose.Lightweight OpenPose uses MobileNet as a feature extraction network,and the prediction layer uses bottleneck-asymmetric structure,thus reducing the amount of the network.The bottleneck-asymmetrical structure compresses the number of input channels of feature maps by 1×1 convolution and replaces the 7×7 convolution structure with the asymmetric structure of 1×7 convolution,7×1 convolution,and 7×7 convolution in parallel.The spatial-temporal graph convolutional network divides the multi-layer convolution into dense blocks,and the convolutional layers in each dense block are connected,thus improving the feature transitivity,enhancing the network’s ability to extract features,thus improving the detection accuracy.Two representative datasets,Multiple Cameras Fall dataset(MCF),and Nanyang Technological University Red Green Blue+Depth Action Recognition dataset(NTU RGB+D),are selected for our experiments,among which NTU RGB+D has two evaluation benchmarks.The results show that the proposed model is superior to the current fall detection models.The accuracy of this network on the MCF dataset is 96.3%,and the accuracies on the two evaluation benchmarks of the NTU RGB+D dataset are 85.6%and 93.5%,respectively.
基金supported by the National Key Research and Development Program of China under Grant 2020YFC2004003 and Grant 2020YFC2004002the National Nature Science Foundation of China(NSFC)under Grant No.61571106.
文摘Masking-based and spectrum mapping-based methods are the two main algorithms of speech enhancement with deep neural network(DNN).But the mapping-based methods only utilizes the phase of noisy speech,which limits the upper bound of speech enhancement performance.Maskingbased methods need to accurately estimate the masking which is still the key problem.Combining the advantages of above two types of methods,this paper proposes the speech enhancement algorithm MM-RDN(maskingmapping residual dense network)based on masking-mapping(MM)and residual dense network(RDN).Using the logarithmic power spectrogram(LPS)of consecutive frames,MM estimates the ideal ratio masking(IRM)matrix of consecutive frames.RDN can make full use of feature maps of all layers.Meanwhile,using the global residual learning to combine the shallow features and deep features,RDN obtains the global dense features from the LPS,thereby improves estimated accuracy of the IRM matrix.Simulations show that the proposed method achieves attractive speech enhancement performance in various acoustic environments.Specifically,in the untrained acoustic test with limited priors,e.g.,unmatched signal-to-noise ratio(SNR)and unmatched noise category,MM-RDN can still outperform the existing convolutional recurrent network(CRN)method in themeasures of perceptual evaluation of speech quality(PESQ)and other evaluation indexes.It indicates that the proposed algorithm is more generalized in untrained conditions.
文摘In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of each layer were enhanced into the global features by the proposed residual dense block,which ensured that the generated images retain more details.Secondly,a multi-scale structure was adopted to extract multi-scale semantic features of the original images.Finally,the features obtained from the dual channels were fused by an adaptive fusion module to further optimize the features.The discriminant network adopted the structure of the Markov discriminator.In addition,by constructing mean square error,structural similarity,and perceived color loss function,the generated image is consistent with the reference image in structure,color,and content.The experimental results showed that the enhanced underwater image deblurring effect of the proposed algorithm was good and the problem of underwater image color bias was effectively improved.In both subjective and objective evaluation indexes,the experimental results of the proposed algorithm are better than those of the comparison algorithm.
基金supported in part by the National 973Program of China(No.2013CB329205)the National Science Foundation of China(No.61401032)
文摘Visible light positioning (VLP) is an emerging candidate for indoor positioning, which carl simultaneously meet the requirements for accuracy, cost, coverage area, and security. However, intercell interference causcd by light intensity superposition linfits the application of VLP. In this Letter, we propose a united block sequence mapping (UBSM)-based VLP that utilizes superposition to integrate the multidimensional information from dense small cells into 2D information. The experimental result shows that UBSM-based VLP can achieve an accuracy of 1.5 cm with a 0.4 m row spacing and 0.35 m column spacing of LED lights.
文摘WALKING on the streets of Jingrong Town in Chengdu,capital of Sichuan Province,bustling groups of people gather around the blocks of businesses there.This dense entrepreneurial climate sharply contrasts with the scene one year ago.Ghost Town Revitalized More than 2,000 years ago,Li Bing,chief of Shu Prefecture(roughly modern-day Sichuan),and his son built a water conservancy project in Dujiangyan,a small city in Sichuan Province.
基金the National Natural Science Foundation of China(No.81830052)the Shanghai Natural Science Foundation of China(No.20ZR1438300)the Shanghai Science and Technology Support Project(No.18441900500),China。
文摘To overcome the computational burden of processing three-dimensional(3 D)medical scans and the lack of spatial information in two-dimensional(2 D)medical scans,a novel segmentation method was proposed that integrates the segmentation results of three densely connected 2 D convolutional neural networks(2 D-CNNs).In order to combine the lowlevel features and high-level features,we added densely connected blocks in the network structure design so that the low-level features will not be missed as the network layer increases during the learning process.Further,in order to resolve the problems of the blurred boundary of the glioma edema area,we superimposed and fused the T2-weighted fluid-attenuated inversion recovery(FLAIR)modal image and the T2-weighted(T2)modal image to enhance the edema section.For the loss function of network training,we improved the cross-entropy loss function to effectively avoid network over-fitting.On the Multimodal Brain Tumor Image Segmentation Challenge(BraTS)datasets,our method achieves dice similarity coefficient values of 0.84,0.82,and 0.83 on the BraTS2018 training;0.82,0.85,and 0.83 on the BraTS2018 validation;and 0.81,0.78,and 0.83 on the BraTS2013 testing in terms of whole tumors,tumor cores,and enhancing cores,respectively.Experimental results showed that the proposed method achieved promising accuracy and fast processing,demonstrating good potential for clinical medicine.
文摘Many networks are designed to stack a large number of residual blocks,deepen the network and improve network performance through short residual connec-tion,long residual connection,and dense connection.However,without consider-ing different contributions of different depth features to the network,these de-signs have the problem of evaluating the importance of different depth features.To solve this problem,this paper proposes an adaptive densely residual net-work(ADRNet)for the single image super resolution.ADRN realizes the evalua-tion of distributions of different depth features and learns more representative features.An adaptive densely residual block(ADRB)was designed,combining 3 residual blocks(RB)and dense connection was added.It learned the attention score of each dense connection through adaptive dense connections,and the at-tention score reflected the importance of the features of each RB.To further en-hance the performance of ADRB,a multi-direction attention block(MDAB)was introduced to obtain multidirectional context information.Through comparative experiments,it is proved that theproposed ADRNet is superior to the existing methods.Through ablation experiments,it is proved that evaluating features of different depths helps to improve network performance.
文摘滚动轴承是风电机组关键部件,其运行工况复杂,故障类型难以准确识别。针对传统深度神经网络在强噪声环境下特征学习能力不足的问题,提出一种基于稠密连接模块的改进深度残差收缩网络(Deep residual shrinkage network based on dense block,DB-DRSN),实现强噪声、不同负载工况下滚动轴承故障的高效诊断。首先,将添加不同等级噪声的振动信号间隔采样并矩阵化,构建二维灰度图作为输入样本。然后,基于Dense block构造稠密连接残差收缩模块层(Residual shrinkage block unit based on dense block,DB-RSBU),利用Bottleneck层替代残差收缩模块中的卷积隐层,并加入Concat连接,达到对浅层和深层特征的充分利用。在每次稠密连接后通过1×1卷积进行降维,利用注意力模块和软阈值对逐通道特征赋不同阈值并降噪。最后,输入样本经过卷积池化层和DB-RSBU层堆叠的网络得到分类结果。试验表明,DB-DRSN模型在CWRU与PU滚动轴承数据集上不同噪声等级下的平均诊断准确率分别达到99.80%和96.44%,相比其他模型有更高的准确率、更快的收敛速度和更强的抗干扰能力。引入稠密连接核心思想对网络结构的改进可为基于数据驱动的风电机组滚动轴承故障诊断方法提供新思路。
基金This work was supported in part by the Basic Scientific Research Project of Liaoning Provincial Department of Education under Grant Nos.LJKQZ2021152 and LJ2020JCL007in part by the National Science Foundation of China(NSFC)under Grant No.61602226in part by the PhD Startup Foundation of Liaoning Technical University of China under Grant Nos.18-1021.
文摘To address the problems of lack of high-frequency information and texture details and unstable training in superresolution generative adversarial net-works,this paper optimizes the generator and discriminator based on the SRGAN model.First,the residual dense block is used as the basic structural unit of the gen-erator to improve the network’s feature extraction capability.Second,enhanced lightweight coordinate attention is incorporated to help the network more precisely concentrate on high-frequency location information,thereby allowing the gener-ator to produce more realistic image reconstruction results.Then,we propose a symmetric and efficient pyramidal segmentation attention discriminator network in which the attention mechanism is capable of derivingfiner-grained multiscale spatial information and creating long-term dependencies between multiscale chan-nel attentions,thus enhancing the discriminative ability of the network.Finally,a Charbonnier loss function and a gradient variance loss function with improved robustness are used to better realize the image’s texture structure and enhance the model’s stability.Thefindings from the experiments reveal that the reconstructed image quality enhances the average peak signal-to-noise ratio(PSNR)by 1.59 dB and the structural similarity index(SSIM)by 0.045 when compared to SRGAN on the three test sets.Compared with the state-of-the-art methods,the reconstructed images have a clearer texture structure,richer high-frequency details,and better visual effects.
基金supported in part by the Basic Scientific Research Project of Liaoning Provincial Department of Education under Grant No.LJKQZ2021152in part by the National Science Foundation of China (NSFC)under Grant No.61602226in part by the PhD Startup Foundation of Liaoning Technical University of China under Grant No.18-1021.
文摘The superresolution(SR)method based on generative adversarial networks(GANs)cannot adequately capture enough diversity from training data,resulting in misalignment between input low resolution(LR)images and output high resolution(HR)images.GAN training has difficulty converging.Based on this,an advanced GAN-based image SR reconstructionmethod is presented.First,the dense connection residual block and attention mechanism are integrated into the GAN generator to improve high-frequency feature extraction.Meanwhile,an added discriminator is added into the GAN discriminant network,which forms a dual discriminator to ensure that the process of training is stable.Second,the more robust Charbonnier loss is used instead of the mean square error(MSE)loss to compare similarities between the obtained image and actual image,and the total variation(TV)loss is employed to smooth the training results.Finally,the experimental results indicate that global structures can be better reconstructed using the method of this paper and texture details of images compared with other SOTA methods.The peak signal-to-noise ratio(PSNR)values by the method of this paper are improved by an average of 2.24 dB,and the structural similarity index measure(SSIM)values are improved by an average of 0.07.