In the coal mining industry,the gangue separation phase imposes a key challenge due to the high visual similaritybetween coal and gangue.Recently,separation methods have become more intelligent and efficient,using new...In the coal mining industry,the gangue separation phase imposes a key challenge due to the high visual similaritybetween coal and gangue.Recently,separation methods have become more intelligent and efficient,using newtechnologies and applying different features for recognition.One such method exploits the difference in substancedensity,leading to excellent coal/gangue recognition.Therefore,this study uses density differences to distinguishcoal from gangue by performing volume prediction on the samples.Our training samples maintain a record of3-side images as input,volume,and weight as the ground truth for the classification.The prediction process relieson a Convolutional neural network(CGVP-CNN)model that receives an input of a 3-side image and then extractsthe needed features to estimate an approximation for the volume.The classification was comparatively performedvia ten different classifiers,namely,K-Nearest Neighbors(KNN),Linear Support Vector Machines(Linear SVM),Radial Basis Function(RBF)SVM,Gaussian Process,Decision Tree,Random Forest,Multi-Layer Perceptron(MLP),Adaptive Boosting(AdaBosst),Naive Bayes,and Quadratic Discriminant Analysis(QDA).After severalexperiments on testing and training data,results yield a classification accuracy of 100%,92%,95%,96%,100%,100%,100%,96%,81%,and 92%,respectively.The test reveals the best timing with KNN,which maintained anaccuracy level of 100%.Assessing themodel generalization capability to newdata is essential to ensure the efficiencyof the model,so by applying a cross-validation experiment,the model generalization was measured.The useddataset was isolated based on the volume values to ensure the model generalization not only on new images of thesame volume but with a volume outside the trained range.Then,the predicted volume values were passed to theclassifiers group,where classification reported accuracy was found to be(100%,100%,100%,98%,88%,87%,100%,87%,97%,100%),respectively.Although obtaining a classification with high accuracy is the main motive,this workhas a remarkable reduction in the data preprocessing time compared to related works.The CGVP-CNN modelmanaged to reduce the data preprocessing time of previous works to 0.017 s while maintaining high classificationaccuracy using the estimated volume value.展开更多
Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware reso...Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.展开更多
Background The use of remote photoplethysmography(rPPG)to estimate blood volume pulse in a noncontact manner has been an active research topic in recent years.Existing methods are primarily based on a singlescale regi...Background The use of remote photoplethysmography(rPPG)to estimate blood volume pulse in a noncontact manner has been an active research topic in recent years.Existing methods are primarily based on a singlescale region of interest(ROI).However,some noise signals that are not easily separated in a single-scale space can be easily separated in a multi-scale space.Also,existing spatiotemporal networks mainly focus on local spatiotemporal information and do not emphasize temporal information,which is crucial in pulse extraction problems,resulting in insufficient spatiotemporal feature modelling.Methods Here,we propose a multi-scale facial video pulse extraction network based on separable spatiotemporal convolution(SSTC)and dimension separable attention(DSAT).First,to solve the problem of a single-scale ROI,we constructed a multi-scale feature space for initial signal separation.Second,SSTC and DSAT were designed for efficient spatiotemporal correlation modeling,which increased the information interaction between the long-span time and space dimensions;this placed more emphasis on temporal features.Results The signal-to-noise ratio(SNR)of the proposed network reached 9.58dB on the PURE dataset and 6.77dB on the UBFC-rPPG dataset,outperforming state-of-the-art algorithms.Conclusions The results showed that fusing multi-scale signals yielded better results than methods based on only single-scale signals.The proposed SSTC and dimension-separable attention mechanism will contribute to more accurate pulse signal extraction.展开更多
Automatic modulation recognition(AMR)of radiation source signals is a research focus in the field of cognitive radio.However,the AMR of radiation source signals at low SNRs still faces a great challenge.Therefore,the ...Automatic modulation recognition(AMR)of radiation source signals is a research focus in the field of cognitive radio.However,the AMR of radiation source signals at low SNRs still faces a great challenge.Therefore,the AMR method of radiation source signals based on two-dimensional data matrix and improved residual neural network is proposed in this paper.First,the time series of the radiation source signals are reconstructed into two-dimensional data matrix,which greatly simplifies the signal preprocessing process.Second,the depthwise convolution and large-size convolutional kernels based residual neural network(DLRNet)is proposed to improve the feature extraction capability of the AMR model.Finally,the model performs feature extraction and classification on the two-dimensional data matrix to obtain the recognition vector that represents the signal modulation type.Theoretical analysis and simulation results show that the AMR method based on two-dimensional data matrix and improved residual network can significantly improve the accuracy of the AMR method.The recognition accuracy of the proposed method maintains a high level greater than 90% even at -14 dB SNR.展开更多
Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on ...Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on the pruning of standard convolutional networks,and they rely intensively on time-consuming fine-tuning to achieve the performance improvement.To this end,we present a novel efficient probability-based channel pruning method for depthwise separable convolutional networks.Our method leverages a new simple yet effective probability-based channel pruning criterion by taking the scaling and shifting factors of batch normalization layers into consideration.A novel shifting factor fusion technique is further developed to improve the performance of the pruned networks without requiring extra time-consuming fine-tuning.We apply the proposed method to five representative deep learning networks,namely MobileNetV1,MobileNetV2,ShuffleNetV1,ShuffleNetV2,and GhostNet,to demonstrate the efficiency of our pruning method.Extensive experimental results and comparisons on publicly available CIFAR10,CIFAR100,and ImageNet datasets validate the feasibility of the proposed method.展开更多
Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor ...Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor one-resistor(1T1R)memristor arrays is limited by the non-ideality of the devices,which prevents the hardware implementation of large and complex networks.In this work,we propose the depthwise separable convolution and bidirectional gate recurrent unit(DSC-BiGRU)network,a lightweight and highly robust hybrid neural network based on 1T1R arrays that enables efficient processing of EEG signals in the temporal,frequency and spatial domains by hybridizing DSC and BiGRU blocks.The network size is reduced and the network robustness is improved while ensuring the network classification accuracy.In the simulation,the measured non-idealities of the 1T1R array are brought into the network through statistical analysis.Compared with traditional convolutional networks,the network parameters are reduced by 95%and the network classification accuracy is improved by 21%at a 95%array yield rate and 5%tolerable error.This work demonstrates that lightweight and highly robust networks based on memristor arrays hold great promise for applications that rely on low consumption and high efficiency.展开更多
Deep kernel mapping support vector machines have achieved good results in numerous tasks by mapping features from a low-dimensional space to a high-dimensional space and then using support vector machines for classifi...Deep kernel mapping support vector machines have achieved good results in numerous tasks by mapping features from a low-dimensional space to a high-dimensional space and then using support vector machines for classification.However,the depth kernel mapping support vector machine does not take into account the connection of different dimensional spaces and increases the model parameters.To further improve the recognition capability of deep kernel mapping support vector machines while reducing the number of model parameters,this paper proposes a framework of Lightweight Deep Convolutional Cross-Connected Kernel Mapping Support Vector Machines(LC-CKMSVM).The framework consists of a feature extraction module and a classification module.The feature extraction module first maps the data from low-dimensional to high-dimensional space by fusing the representations of different dimensional spaces through cross-connections;then,it uses depthwise separable convolution to replace part of the original convolution to reduce the number of parameters in the module;The classification module uses a soft margin support vector machine for classification.The results on 6 different visual datasets show that LC-CKMSVM obtains better classification accuracies on most cases than the other five models.展开更多
Traditional separation methods have limited ability to handle the speech separation problem in high reverberant and low signal-to-noise ratio(SNR)environments,and thus achieve unsatisfactory results.In this study,a co...Traditional separation methods have limited ability to handle the speech separation problem in high reverberant and low signal-to-noise ratio(SNR)environments,and thus achieve unsatisfactory results.In this study,a convolutional neural network with temporal convolution and residual network(TC-ResNet)is proposed to realize speech separation in a complex acoustic environment.A simplified steered-response power phase transform,denoted as GSRP-PHAT,is employed to reduce the computational cost.The extracted features are reshaped to a special tensor as the system inputs and implements temporal convolution,which not only enlarges the receptive field of the convolution layer but also significantly reduces the network computational cost.Residual blocks are used to combine multiresolution features and accelerate the training procedure.A modified ideal ratio mask is applied as the training target.Simulation results demonstrate that the proposed microphone array speech separation algorithm based on TC-ResNet achieves a better performance in terms of distortion ratio,source-to-interference ratio,and short-time objective intelligibility in low SNR and high reverberant environments,particularly in untrained situations.This indicates that the proposed method has generalization to untrained conditions.展开更多
基金the National Natural Science Foundation of China under Grant No.52274159 received by E.Hu,https://www.nsfc.gov.cn/Grant No.52374165 received by E.Hu,https://www.nsfc.gov.cn/the China National Coal Group Key Technology Project Grant No.(20221CY001)received by Z.Guan,and E.Hu,https://www.chinacoal.com/.
文摘In the coal mining industry,the gangue separation phase imposes a key challenge due to the high visual similaritybetween coal and gangue.Recently,separation methods have become more intelligent and efficient,using newtechnologies and applying different features for recognition.One such method exploits the difference in substancedensity,leading to excellent coal/gangue recognition.Therefore,this study uses density differences to distinguishcoal from gangue by performing volume prediction on the samples.Our training samples maintain a record of3-side images as input,volume,and weight as the ground truth for the classification.The prediction process relieson a Convolutional neural network(CGVP-CNN)model that receives an input of a 3-side image and then extractsthe needed features to estimate an approximation for the volume.The classification was comparatively performedvia ten different classifiers,namely,K-Nearest Neighbors(KNN),Linear Support Vector Machines(Linear SVM),Radial Basis Function(RBF)SVM,Gaussian Process,Decision Tree,Random Forest,Multi-Layer Perceptron(MLP),Adaptive Boosting(AdaBosst),Naive Bayes,and Quadratic Discriminant Analysis(QDA).After severalexperiments on testing and training data,results yield a classification accuracy of 100%,92%,95%,96%,100%,100%,100%,96%,81%,and 92%,respectively.The test reveals the best timing with KNN,which maintained anaccuracy level of 100%.Assessing themodel generalization capability to newdata is essential to ensure the efficiencyof the model,so by applying a cross-validation experiment,the model generalization was measured.The useddataset was isolated based on the volume values to ensure the model generalization not only on new images of thesame volume but with a volume outside the trained range.Then,the predicted volume values were passed to theclassifiers group,where classification reported accuracy was found to be(100%,100%,100%,98%,88%,87%,100%,87%,97%,100%),respectively.Although obtaining a classification with high accuracy is the main motive,this workhas a remarkable reduction in the data preprocessing time compared to related works.The CGVP-CNN modelmanaged to reduce the data preprocessing time of previous works to 0.017 s while maintaining high classificationaccuracy using the estimated volume value.
文摘Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.
基金Supported by the National Natural Science Foundation of China(61903336,61976190)the Natural Science Foundation of Zhejiang Province(LY21F030015)。
文摘Background The use of remote photoplethysmography(rPPG)to estimate blood volume pulse in a noncontact manner has been an active research topic in recent years.Existing methods are primarily based on a singlescale region of interest(ROI).However,some noise signals that are not easily separated in a single-scale space can be easily separated in a multi-scale space.Also,existing spatiotemporal networks mainly focus on local spatiotemporal information and do not emphasize temporal information,which is crucial in pulse extraction problems,resulting in insufficient spatiotemporal feature modelling.Methods Here,we propose a multi-scale facial video pulse extraction network based on separable spatiotemporal convolution(SSTC)and dimension separable attention(DSAT).First,to solve the problem of a single-scale ROI,we constructed a multi-scale feature space for initial signal separation.Second,SSTC and DSAT were designed for efficient spatiotemporal correlation modeling,which increased the information interaction between the long-span time and space dimensions;this placed more emphasis on temporal features.Results The signal-to-noise ratio(SNR)of the proposed network reached 9.58dB on the PURE dataset and 6.77dB on the UBFC-rPPG dataset,outperforming state-of-the-art algorithms.Conclusions The results showed that fusing multi-scale signals yielded better results than methods based on only single-scale signals.The proposed SSTC and dimension-separable attention mechanism will contribute to more accurate pulse signal extraction.
基金National Natural Science Foundation of China under Grant No.61973037China Postdoctoral Science Foundation under Grant No.2022M720419。
文摘Automatic modulation recognition(AMR)of radiation source signals is a research focus in the field of cognitive radio.However,the AMR of radiation source signals at low SNRs still faces a great challenge.Therefore,the AMR method of radiation source signals based on two-dimensional data matrix and improved residual neural network is proposed in this paper.First,the time series of the radiation source signals are reconstructed into two-dimensional data matrix,which greatly simplifies the signal preprocessing process.Second,the depthwise convolution and large-size convolutional kernels based residual neural network(DLRNet)is proposed to improve the feature extraction capability of the AMR model.Finally,the model performs feature extraction and classification on the two-dimensional data matrix to obtain the recognition vector that represents the signal modulation type.Theoretical analysis and simulation results show that the AMR method based on two-dimensional data matrix and improved residual network can significantly improve the accuracy of the AMR method.The recognition accuracy of the proposed method maintains a high level greater than 90% even at -14 dB SNR.
基金the National Natural Science Foundation of China under Grant Nos.62036010 and 62072340the Zhejiang Provincial Natural Science Foundation of China under Grant Nos.LZ21F020001 and LSZ19F020001the Open Project Program of the State Key Laboratory of CAD&CG,Zhejiang University under Grant No.A2220.
文摘Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on the pruning of standard convolutional networks,and they rely intensively on time-consuming fine-tuning to achieve the performance improvement.To this end,we present a novel efficient probability-based channel pruning method for depthwise separable convolutional networks.Our method leverages a new simple yet effective probability-based channel pruning criterion by taking the scaling and shifting factors of batch normalization layers into consideration.A novel shifting factor fusion technique is further developed to improve the performance of the pruned networks without requiring extra time-consuming fine-tuning.We apply the proposed method to five representative deep learning networks,namely MobileNetV1,MobileNetV2,ShuffleNetV1,ShuffleNetV2,and GhostNet,to demonstrate the efficiency of our pruning method.Extensive experimental results and comparisons on publicly available CIFAR10,CIFAR100,and ImageNet datasets validate the feasibility of the proposed method.
基金Project supported by the National Key Research and Development Program of China(Grant No.2019YFB2205102)the National Natural Science Foundation of China(Grant Nos.61974164,62074166,61804181,62004219,62004220,and 62104256).
文摘Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor one-resistor(1T1R)memristor arrays is limited by the non-ideality of the devices,which prevents the hardware implementation of large and complex networks.In this work,we propose the depthwise separable convolution and bidirectional gate recurrent unit(DSC-BiGRU)network,a lightweight and highly robust hybrid neural network based on 1T1R arrays that enables efficient processing of EEG signals in the temporal,frequency and spatial domains by hybridizing DSC and BiGRU blocks.The network size is reduced and the network robustness is improved while ensuring the network classification accuracy.In the simulation,the measured non-idealities of the 1T1R array are brought into the network through statistical analysis.Compared with traditional convolutional networks,the network parameters are reduced by 95%and the network classification accuracy is improved by 21%at a 95%array yield rate and 5%tolerable error.This work demonstrates that lightweight and highly robust networks based on memristor arrays hold great promise for applications that rely on low consumption and high efficiency.
基金This work is supported by the National Natural Science Foundation of China(61806013,61876010,61906005,62166002)General project of Science and Technology Plan of Beijing Municipal Education Commission(KM202110005028)+1 种基金Project of Interdisciplinary Research Institute of Beijing University of Technology(2021020101)International Research Cooperation Seed Fund of Beijing University of Technology(2021A01).
文摘Deep kernel mapping support vector machines have achieved good results in numerous tasks by mapping features from a low-dimensional space to a high-dimensional space and then using support vector machines for classification.However,the depth kernel mapping support vector machine does not take into account the connection of different dimensional spaces and increases the model parameters.To further improve the recognition capability of deep kernel mapping support vector machines while reducing the number of model parameters,this paper proposes a framework of Lightweight Deep Convolutional Cross-Connected Kernel Mapping Support Vector Machines(LC-CKMSVM).The framework consists of a feature extraction module and a classification module.The feature extraction module first maps the data from low-dimensional to high-dimensional space by fusing the representations of different dimensional spaces through cross-connections;then,it uses depthwise separable convolution to replace part of the original convolution to reduce the number of parameters in the module;The classification module uses a soft margin support vector machine for classification.The results on 6 different visual datasets show that LC-CKMSVM obtains better classification accuracies on most cases than the other five models.
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Traditional separation methods have limited ability to handle the speech separation problem in high reverberant and low signal-to-noise ratio(SNR)environments,and thus achieve unsatisfactory results.In this study,a convolutional neural network with temporal convolution and residual network(TC-ResNet)is proposed to realize speech separation in a complex acoustic environment.A simplified steered-response power phase transform,denoted as GSRP-PHAT,is employed to reduce the computational cost.The extracted features are reshaped to a special tensor as the system inputs and implements temporal convolution,which not only enlarges the receptive field of the convolution layer but also significantly reduces the network computational cost.Residual blocks are used to combine multiresolution features and accelerate the training procedure.A modified ideal ratio mask is applied as the training target.Simulation results demonstrate that the proposed microphone array speech separation algorithm based on TC-ResNet achieves a better performance in terms of distortion ratio,source-to-interference ratio,and short-time objective intelligibility in low SNR and high reverberant environments,particularly in untrained situations.This indicates that the proposed method has generalization to untrained conditions.
