Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the...Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the deep learning approach have been proposed for activity recognition purpose.However,with the increasing advancements in technologies having limited computational resources,it needs to design an efficient deep learning-based approaches with improved utilization of computational resources.This paper presents a simple and efficient 2-dimensional CNN(2-D CNN)architecture with very small-size convolutional kernel for human activity recognition.The merit of the proposed CNN architecture over standard deep learning architectures is fewer trainable parameters and lesser memory requirement which enables it to train the proposed CNN architecture on low GPU memory-based devices and also works well with smaller as well as larger size datasets.The proposed approach consists of mainly four stages:namely(1)creation of dataset and data augmentation,(2)designing 2-D CNN architecture,(3)the proposed 2-D CNN architecture trained from scratch up to optimum stage,and(4)evaluation of the trained 2-D CNN architecture.To illustrate the effectiveness of the proposed architecture several extensive experiments are conducted on three publicly available datasets,namely IXMAS,YouTube,and UCF101 dataset.The results of the proposed method and its comparison with other state-of-the-art methods demonstrate the usefulness of the proposed method.展开更多
Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases...Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases.However,DL architectures are computationally more demanding.In recent years,researchers have focused on combining the computationally less intensive portion of the DL architectures with ML approaches,say for example,combining the convolutional layer blocks of Convolution Neural Networks(CNNs)into ML algorithms such as Extreme Gradient Boosting(XGBoost)and K-Nearest Neighbor(KNN)resulting in CNN-XGBoost and CNN-KNN,respectively.However,these approaches are homogenous in the sense that they use a fixed Activation Function(AFs)in the sequence of convolution and pooling layers,thereby limiting the ability to capture unique features.Since various AFs are readily available and each could capture unique features,we propose a Convolutionbased Heterogeneous Activation Facility(CHAF)which uses multiple AFs in the convolution layer blocks,one for each block,with a motivation of extracting features in a better manner to improve the accuracy.The proposed CHAF approach is validated on PTB and shown to outperform the homogeneous approaches such as CNN-KNN and CNN-XGBoost.For PTB dataset,proposed CHAF-KNN has an accuracy of 99.55%and an F1 score of 99.68%in just 0.008 s,outperforming the state-of-the-art CNN-XGBoost which has an accuracy of 99.38%and an F1 score of 99.32%in 1.23 s.To validate the generality of the proposed CHAF,experiments were repeated on MIT-BIH dataset,and the proposed CHAF-KNN is shown to outperform CNN-KNN and CNN-XGBoost.展开更多
Convolutional neural networks （CNNs） have been applied in state-of-the-art visual tracking tasks to represent the target. However, most existing algorithms treat visual tracking as an object-specific task. Therefore...Convolutional neural networks （CNNs） have been applied in state-of-the-art visual tracking tasks to represent the target. However, most existing algorithms treat visual tracking as an object-specific task. Therefore, the model needs to be retrained for different test video sequences. We propose a branch-activated multi-domain convolutional neural network （BAMDCNN）. In contrast to most existing trackers based on CNNs which require frequent online training, BAMDCNN only needs offine training and online fine-tuning. Specifically, BAMDCNN exploits category-specific features that are more robust against variations. To allow for learning category-specific information, we introduce a group algorithm and a branch activation method. Experimental results on challenging benchmark show that the proposed algorithm outperforms other state-of-the-art methods. What＇s more, compared with CNN based trackers, BAMDCNN increases tracking speed.展开更多
With the improvement of people’s living standards,the demand for health monitoring and exercise detection is increasing.It is of great significance to study human activity recognition(HAR)methods that are different f...With the improvement of people’s living standards,the demand for health monitoring and exercise detection is increasing.It is of great significance to study human activity recognition(HAR)methods that are different from traditional feature extraction methods.This article uses convolutional neural network(CNN)algorithms in deep learning to automatically extract features of activities related to human life.We used a stochastic gradient descent algorithm to optimize the parameters of the CNN.The trained network model is compressed on STM32CubeMX-AI.Finally,this article introduces the use of neural networks on embedded devices to recognize six human activities of daily life,such as sitting,standing,walking,jogging,upstairs,and downstairs.The acceleration sensor related to human activity information is used to obtain the relevant characteristics of the activity,thereby solving the HAR problem.By drawing the accuracy curve,loss function curve,and confusion matrix diagram of the training model,the recognition effect of the convolutional neural network can be seen more intuitively.After comparing the average accuracy of each set of experiments and the test set of the best model obtained from it,the best model is then selected.展开更多
Hydrogen evolution reaction(HER)plays a crucial role in developing clean and renewable hydrogen energy technologies.However,conventional HER catalysts rely on expensive and scarce noble metals,which is a significant c...Hydrogen evolution reaction(HER)plays a crucial role in developing clean and renewable hydrogen energy technologies.However,conventional HER catalysts rely on expensive and scarce noble metals,which is a significant challenge for practical application.Recently,twodimensional transition metal dichalcogenides(2D-TMDs)have emerged as attractive and cost-effective alternatives for efficient electrocatalysis in the HER.Substantial efforts have been dedicated to advancing the synthesis and application of 2D-TMDs.This review highlights the design and synthesis of high-performance 2D-TMDs-based HER electrocatalysts by combining theoretical calculations with experimental methods.Subsequently,recent advances in synthesizing different types of 2D TMDs with enhanced HER activity are summarized.Finally,the conclusion and perspectives of the 2D TMDs-based HER electrocatalysts are discussed.We expect that this review will provide new insights into the design and development of highly efficient 2D TMDs-based HER electrocatalysts for industrial applications.展开更多
This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophagea...This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.展开更多
In this paper, an improved two-dimensional convolution neural network(2DCNN) is proposed to monitor and analyze elevator health, based on the distribution characteristics of elevator time series data in two-dimensiona...In this paper, an improved two-dimensional convolution neural network(2DCNN) is proposed to monitor and analyze elevator health, based on the distribution characteristics of elevator time series data in two-dimensional images. The current and effective power signals from an elevator traction machine are collected to generate gray-scale binary images. The improved two-dimensional convolution neural network is used to extract deep features from the images for classification, so as to recognize the elevator working conditions. Furthermore, the oscillation criterion is proposed to describe and analyze the active power oscillations. The current and active power are used to synchronously describe the working condition of the elevator, which can explain the co-occurrence state and potential relationship of elevator data. Based on the improved integration of local features of the time series, the recognition accuracy of the proposed 2DCNN is 97.78%, which is better than that of a one-dimensional convolution neural network. This research can improve the real-time monitoring and visual analysis performance of the elevator maintenance personnel, as well as improve their work efficiency.展开更多
Two-dimensional elec tronic spec troscopy(2DES)is a powerful met hod to probe the coherent electron dynamics in complicated systems.Stabilizing the phase difference of the incident ultrashort pulses is the mos t chall...Two-dimensional elec tronic spec troscopy(2DES)is a powerful met hod to probe the coherent electron dynamics in complicated systems.Stabilizing the phase difference of the incident ultrashort pulses is the mos t challenging par t for experimen tal demonstration of 2DES.Here,we present a tuto rial review on the 2DES proto cols based on active phase managements which are originally developed for quantum optics experiments.We introduce the 2DES techniques in box and pump-probe geometries with phase stabilization realized by interferometry,and outline the fully collinear 2DES approach with the frequency tagging by acoustic optical modulators and frequency combs.The combination of active phase managements,ultrashort pulses and other spectroscopic methods may open new opportunities to tackle essential challenges related to excited states.展开更多
Lanthanum modified materials have been widely used for the removal of hazardous anions.In this study,in situ ATR-FTIR and two-dimensional correlation analysis were employed to investigate the adsorption mechanism of a...Lanthanum modified materials have been widely used for the removal of hazardous anions.In this study,in situ ATR-FTIR and two-dimensional correlation analysis were employed to investigate the adsorption mechanism of arsenate（As（V）） on lanthanum-impregnated activated alumina（LAA）.Our results showed that electrostatic interaction attracted As（V） anions to the LAA surface,and then As（V） could form monodentate configuration on the LAA surface at pH 5-9.The result of 2D-COS showed that two coexistent adsorbed As（V） species,H2AsO4^- and HAsO4^2-,were adsorbed on the LAA surface without specific sequence at different pH conditions,indicating a negligible role of the incorporated protons of As（V） on the adsorption affinity to LAA surface.The results of this study reveal insights into LAA surface complexes on the molecular scale and provide theoretical support to new metal oxides design for efficient arsenic removal.展开更多
Hydrogen energy could be a economic and powerful technology for sustainable future. Producing hydrogen fuel by electrochemical water splitting has attracted intense interest. Due to their physical and chemical propert...Hydrogen energy could be a economic and powerful technology for sustainable future. Producing hydrogen fuel by electrochemical water splitting has attracted intense interest. Due to their physical and chemical properties, two-dimensional(2 D) nanomaterials have sparked immense interest in water electrocatalysis for hydrogen production. This review focuses on the emerging nanocatalysts in 2 D nanoarchitectures for electrocatalytic hydrogen production. The fundamentals of HER are firstly depicted, following the discussion of recent advances in typical 2 D electrocatalysts for HER. The insights into the relationship among the synthetic protocols, structure, catalytic performance and thermodynamics will be discussed in details. Finally, the outlooks regarding further development of 2 D nanocatalysts for HER are proposed.We hope this review will offer a comprehensive understanding in 2 D nanocatalysts to promote electrochemical hydrogen production.展开更多
Laser-induced breakdown spectroscopy(LIBS)has been applied to many fields for the quantitative analysis of diverse materials.Improving the prediction accuracy of LIBS regression models is still of great significance f...Laser-induced breakdown spectroscopy(LIBS)has been applied to many fields for the quantitative analysis of diverse materials.Improving the prediction accuracy of LIBS regression models is still of great significance for the Mars exploration in the near future.In this study,we explored the quantitative analysis of LIBS for the one-dimensional Chem Cam(an instrument containing a LIBS spectrometer and a Remote Micro-Imager)spectral data whose spectra are produced by the Chem Cam team using LIBS under the Mars-like atmospheric conditions.We constructed a convolutional neural network(CNN)regression model with unified parameters for all oxides,which is efficient and concise.CNN that has the excellent capability of feature extraction can effectively overcome the chemical matrix effects that impede the prediction accuracy of regression models.Firstly,we explored the effects of four activation functions on the performance of the CNN model.The results show that the CNN model with the hyperbolic tangent(tanh)function outperforms the CNN models with the other activation functions(the rectified linear unit function,the linear function and the Sigmoid function).Secondly,we compared the performance among the CNN models using different optimization methods.The CNN model with the stochastic gradient descent optimization and the initial learning rate?=?0.0005 achieves satisfactory performance compared to the other CNN models.Finally,we compared the performance of the CNN model,the model based on support vector regression(SVR)and the model based on partial least square regression(PLSR).The results exhibit the CNN model is superior to the SVR model and the PLSR model for all oxides.Based on the above analysis,we conclude the CNN regression model can effectively improve the prediction accuracy of LIBS.展开更多
A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically ge...A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically generated TGFROG traces to complete supervised trainings of the convolutional neural networks,then use similarly generated traces not included in the training dataset to test how well the networks are trained.Accurate retrieval of such traces by the neural network is realized.In our case,we find that networks with exponential linear unit(ELU) activation function perform better than those with leaky rectified linear unit(LRELU) and scaled exponential linear unit(SELU).Finally,the issues that need to be addressed for the retrieval of experimental data by this method are discussed.展开更多
With the continuous development of e-commerce,consumers show increasing interest in posting comments on consumption experience and quality of commodities.Meanwhile,people make purchasing decisions relying on other com...With the continuous development of e-commerce,consumers show increasing interest in posting comments on consumption experience and quality of commodities.Meanwhile,people make purchasing decisions relying on other comments much more than ever before.So the reliability of commodity comments has a significant impact on ensuring consumers’equity and building a fair internet-trade-environment.However,some unscrupulous online-sellers write fake praiseful reviews for themselves and malicious comments for their business counterparts to maximize their profits.Those improper ways of self-profiting have severely ruined the entire online shopping industry.Aiming to detect and prevent these deceptive comments effectively,we construct a model of Multi-Filters Convolutional Neural Network(MFCNN)for opinion spam detection.MFCNN is designed with a fixed-length sequence input and an improved activation function to avoid the gradient vanishing problem in spam opinion detection.Moreover,convolution filters with different widths are used in MFCNN to represent the sentences and documents.Our experimental results show that MFCNN outperforms current state-of-the-art methods on standard spam detection benchmarks.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
Traditional methods for judging the degree of insufficient fluid supply in oil wells have low efficiency and limited accuracy. To address this problem, a method for intelligently identifying the degree of insufficient...Traditional methods for judging the degree of insufficient fluid supply in oil wells have low efficiency and limited accuracy. To address this problem, a method for intelligently identifying the degree of insufficient fluid supply in oil wells based on convolutional neural networks is proposed in this paper. Firstly, 5000 indicator diagrams with insufficient liquid supply were collected from the oilfield site, and a sample set was established after preprocessing;then based on the AlexNet model, combined with the characteristics of the indicator diagram, a convolutional neural network model including 4 layers of convolutional layers, 3 layers of down-pooling layers and 2 layers of fully connected layers is established. The backpropagation, ReLu activation function and dropout regularization method are used to complete the training of the convolutional neural network;finally, the performance of the convolutional neural network under different iteration times and network structure is compared, and the super parameter optimization of the model is completed. It has laid a good foundation for realizing the self-adaptive and intelligent matching of oil well production parameters and formation fluid supply conditions. It has certain application prospects. The results show that the accuracy of training and verification of the method exceeds 98%, which can meet the actual application requirements on site.展开更多
Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of ...Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity.Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process.However,it takes a lot of time and effort for researchers to use the existing activation function in their research.Therefore,in this paper,we propose a universal activation function(UA)so that researchers can easily create and apply various activation functions and improve the performance of neural networks.UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters.The famous Convolutional Neural Network(CNN)and benchmark datasetwere used to evaluate the experimental performance of the UA proposed in this study.We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which theUA is applied.In addition,we evaluated the performance of the new activation function generated by adjusting the hyperparameters of theUA.The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5%through the UA,although most of them showed similar performance to the traditional activation function.展开更多
The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software w...The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two state-of-the-art methods.展开更多
Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automa...Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automation has greatly increased across industries worldwide.Real-time detection requires edge devices with limited computational time.This study proposes a novel hybrid deep learning system for human activity recognition(HAR),aiming to enhance the recognition accuracy and reduce the computational time.The proposed system combines a pretrained image classification model with a sequence analysis model.First,the dataset was divided into a training set(70%),validation set(10%),and test set(20%).Second,all the videos were converted into frames and deep-based features were extracted from each frame using convolutional neural networks(CNNs)with a vision transformer.Following that,bidirectional long short-term memory(BiLSTM)-and temporal convolutional network(TCN)-based models were trained using the training set,and their performances were evaluated using the validation set and test set.Four benchmark datasets(UCF11,UCF50,UCF101,and JHMDB)were used to evaluate the performance of the proposed HAR-based system.The experimental results showed that the combination of ConvNeXt and the TCN-based model achieved a recognition accuracy of 97.73%for UCF11,98.81%for UCF50,98.46%for UCF101,and 83.38%for JHMDB,respectively.This represents improvements in the recognition accuracy of 4%,2.67%,3.67%,and 7.08%for the UCF11,UCF50,UCF101,and JHMDB datasets,respectively,over existing models.Moreover,the proposed HAR-based system obtained superior recognition accuracy,shorter computational times,and minimal memory usage compared to the existing models.展开更多
文摘Recognition of human activity based on convolutional neural network(CNN)has received the interest of researchers in recent years due to its significant improvement in accuracy.A large number of algorithms based on the deep learning approach have been proposed for activity recognition purpose.However,with the increasing advancements in technologies having limited computational resources,it needs to design an efficient deep learning-based approaches with improved utilization of computational resources.This paper presents a simple and efficient 2-dimensional CNN(2-D CNN)architecture with very small-size convolutional kernel for human activity recognition.The merit of the proposed CNN architecture over standard deep learning architectures is fewer trainable parameters and lesser memory requirement which enables it to train the proposed CNN architecture on low GPU memory-based devices and also works well with smaller as well as larger size datasets.The proposed approach consists of mainly four stages:namely(1)creation of dataset and data augmentation,(2)designing 2-D CNN architecture,(3)the proposed 2-D CNN architecture trained from scratch up to optimum stage,and(4)evaluation of the trained 2-D CNN architecture.To illustrate the effectiveness of the proposed architecture several extensive experiments are conducted on three publicly available datasets,namely IXMAS,YouTube,and UCF101 dataset.The results of the proposed method and its comparison with other state-of-the-art methods demonstrate the usefulness of the proposed method.
文摘Machine Learning(ML)and Deep Learning(DL)technologies are revolutionizing the medical domain,especially with Electrocardiogram(ECG),by providing new tools and techniques for diagnosing,treating,and preventing diseases.However,DL architectures are computationally more demanding.In recent years,researchers have focused on combining the computationally less intensive portion of the DL architectures with ML approaches,say for example,combining the convolutional layer blocks of Convolution Neural Networks(CNNs)into ML algorithms such as Extreme Gradient Boosting(XGBoost)and K-Nearest Neighbor(KNN)resulting in CNN-XGBoost and CNN-KNN,respectively.However,these approaches are homogenous in the sense that they use a fixed Activation Function(AFs)in the sequence of convolution and pooling layers,thereby limiting the ability to capture unique features.Since various AFs are readily available and each could capture unique features,we propose a Convolutionbased Heterogeneous Activation Facility(CHAF)which uses multiple AFs in the convolution layer blocks,one for each block,with a motivation of extracting features in a better manner to improve the accuracy.The proposed CHAF approach is validated on PTB and shown to outperform the homogeneous approaches such as CNN-KNN and CNN-XGBoost.For PTB dataset,proposed CHAF-KNN has an accuracy of 99.55%and an F1 score of 99.68%in just 0.008 s,outperforming the state-of-the-art CNN-XGBoost which has an accuracy of 99.38%and an F1 score of 99.32%in 1.23 s.To validate the generality of the proposed CHAF,experiments were repeated on MIT-BIH dataset,and the proposed CHAF-KNN is shown to outperform CNN-KNN and CNN-XGBoost.
基金the Innovation Action Plan Foundation of Shanghai(No.16511101200)
文摘Convolutional neural networks （CNNs） have been applied in state-of-the-art visual tracking tasks to represent the target. However, most existing algorithms treat visual tracking as an object-specific task. Therefore, the model needs to be retrained for different test video sequences. We propose a branch-activated multi-domain convolutional neural network （BAMDCNN）. In contrast to most existing trackers based on CNNs which require frequent online training, BAMDCNN only needs offine training and online fine-tuning. Specifically, BAMDCNN exploits category-specific features that are more robust against variations. To allow for learning category-specific information, we introduce a group algorithm and a branch activation method. Experimental results on challenging benchmark show that the proposed algorithm outperforms other state-of-the-art methods. What＇s more, compared with CNN based trackers, BAMDCNN increases tracking speed.
文摘With the improvement of people’s living standards,the demand for health monitoring and exercise detection is increasing.It is of great significance to study human activity recognition(HAR)methods that are different from traditional feature extraction methods.This article uses convolutional neural network(CNN)algorithms in deep learning to automatically extract features of activities related to human life.We used a stochastic gradient descent algorithm to optimize the parameters of the CNN.The trained network model is compressed on STM32CubeMX-AI.Finally,this article introduces the use of neural networks on embedded devices to recognize six human activities of daily life,such as sitting,standing,walking,jogging,upstairs,and downstairs.The acceleration sensor related to human activity information is used to obtain the relevant characteristics of the activity,thereby solving the HAR problem.By drawing the accuracy curve,loss function curve,and confusion matrix diagram of the training model,the recognition effect of the convolutional neural network can be seen more intuitively.After comparing the average accuracy of each set of experiments and the test set of the best model obtained from it,the best model is then selected.
基金supported by the National Key Projects for Fundamental Research and Development of China(2021YFA1500803)the National Natural Science Foundation of China(51825205,52120105002,22088102,22279150,22209186)+1 种基金the Beijing Natural Science Foundation(2222080)the Youth Innovation Promotion Association of the CAS(Y2021011)。
文摘Hydrogen evolution reaction(HER)plays a crucial role in developing clean and renewable hydrogen energy technologies.However,conventional HER catalysts rely on expensive and scarce noble metals,which is a significant challenge for practical application.Recently,twodimensional transition metal dichalcogenides(2D-TMDs)have emerged as attractive and cost-effective alternatives for efficient electrocatalysis in the HER.Substantial efforts have been dedicated to advancing the synthesis and application of 2D-TMDs.This review highlights the design and synthesis of high-performance 2D-TMDs-based HER electrocatalysts by combining theoretical calculations with experimental methods.Subsequently,recent advances in synthesizing different types of 2D TMDs with enhanced HER activity are summarized.Finally,the conclusion and perspectives of the 2D TMDs-based HER electrocatalysts are discussed.We expect that this review will provide new insights into the design and development of highly efficient 2D TMDs-based HER electrocatalysts for industrial applications.
基金funded by Vietnam National Foundation for Science and Technology Development(NAFOSTED)under grant number:NCUD.02-2024.11.
文摘This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.
基金Sponsored by the National Natural Science Foundation of China (Grant No.61771223)the Key Research and Development Program of Jiangsu Province(Grant No.SBE2018334)。
文摘In this paper, an improved two-dimensional convolution neural network(2DCNN) is proposed to monitor and analyze elevator health, based on the distribution characteristics of elevator time series data in two-dimensional images. The current and effective power signals from an elevator traction machine are collected to generate gray-scale binary images. The improved two-dimensional convolution neural network is used to extract deep features from the images for classification, so as to recognize the elevator working conditions. Furthermore, the oscillation criterion is proposed to describe and analyze the active power oscillations. The current and active power are used to synchronously describe the working condition of the elevator, which can explain the co-occurrence state and potential relationship of elevator data. Based on the improved integration of local features of the time series, the recognition accuracy of the proposed 2DCNN is 97.78%, which is better than that of a one-dimensional convolution neural network. This research can improve the real-time monitoring and visual analysis performance of the elevator maintenance personnel, as well as improve their work efficiency.
基金This work is supported by the National Key R&D Program of China(No.2017YFA0303700 and No.2018YFA0209101)the National Natural Science Foundation of China(No.21922302,No.21873047,No.11904168,No.91833305,and No.91850105)+1 种基金the Priority Academic Program Development of Jiangsu Higher Education Institutionsthe Fundamental Research Funds for the Central University.
文摘Two-dimensional elec tronic spec troscopy(2DES)is a powerful met hod to probe the coherent electron dynamics in complicated systems.Stabilizing the phase difference of the incident ultrashort pulses is the mos t challenging par t for experimen tal demonstration of 2DES.Here,we present a tuto rial review on the 2DES proto cols based on active phase managements which are originally developed for quantum optics experiments.We introduce the 2DES techniques in box and pump-probe geometries with phase stabilization realized by interferometry,and outline the fully collinear 2DES approach with the frequency tagging by acoustic optical modulators and frequency combs.The combination of active phase managements,ultrashort pulses and other spectroscopic methods may open new opportunities to tackle essential challenges related to excited states.
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences(No.XDB14020201)the National Basic Research Program of China(No. 2014CB441102)+1 种基金the National Natural Science Foundation of China(No.21477144)the Young Scientists Fund of RCEES, RCEES-QN-20130017F
文摘Lanthanum modified materials have been widely used for the removal of hazardous anions.In this study,in situ ATR-FTIR and two-dimensional correlation analysis were employed to investigate the adsorption mechanism of arsenate（As（V）） on lanthanum-impregnated activated alumina（LAA）.Our results showed that electrostatic interaction attracted As（V） anions to the LAA surface,and then As（V） could form monodentate configuration on the LAA surface at pH 5-9.The result of 2D-COS showed that two coexistent adsorbed As（V） species,H2AsO4^- and HAsO4^2-,were adsorbed on the LAA surface without specific sequence at different pH conditions,indicating a negligible role of the incorporated protons of As（V） on the adsorption affinity to LAA surface.The results of this study reveal insights into LAA surface complexes on the molecular scale and provide theoretical support to new metal oxides design for efficient arsenic removal.
基金supported by the National Natural Science Foundation of China (Nos. 22075092, 21805103)the Program for HUST Academic Frontier Youth Team (No. 2018QYTD15)+1 种基金the Innovation and Talent Recruitment Base of New Energy Chemistry and Device (No. B21003)the National 1000 Young Talents Program of China。
文摘Hydrogen energy could be a economic and powerful technology for sustainable future. Producing hydrogen fuel by electrochemical water splitting has attracted intense interest. Due to their physical and chemical properties, two-dimensional(2 D) nanomaterials have sparked immense interest in water electrocatalysis for hydrogen production. This review focuses on the emerging nanocatalysts in 2 D nanoarchitectures for electrocatalytic hydrogen production. The fundamentals of HER are firstly depicted, following the discussion of recent advances in typical 2 D electrocatalysts for HER. The insights into the relationship among the synthetic protocols, structure, catalytic performance and thermodynamics will be discussed in details. Finally, the outlooks regarding further development of 2 D nanocatalysts for HER are proposed.We hope this review will offer a comprehensive understanding in 2 D nanocatalysts to promote electrochemical hydrogen production.
基金supported by the Pre-research project on Civil Aerospace Technologies(No.D020102)funded by China National Space Administration(CNSA)the funding from National Natural Science Foundation of China(Nos.U1931211,41573056)+1 种基金the Natural Science Foundation of Shandong Province(No.ZR2019MD008)the Major Research Project of Shandong Province(No.GG201809130208)。
文摘Laser-induced breakdown spectroscopy(LIBS)has been applied to many fields for the quantitative analysis of diverse materials.Improving the prediction accuracy of LIBS regression models is still of great significance for the Mars exploration in the near future.In this study,we explored the quantitative analysis of LIBS for the one-dimensional Chem Cam(an instrument containing a LIBS spectrometer and a Remote Micro-Imager)spectral data whose spectra are produced by the Chem Cam team using LIBS under the Mars-like atmospheric conditions.We constructed a convolutional neural network(CNN)regression model with unified parameters for all oxides,which is efficient and concise.CNN that has the excellent capability of feature extraction can effectively overcome the chemical matrix effects that impede the prediction accuracy of regression models.Firstly,we explored the effects of four activation functions on the performance of the CNN model.The results show that the CNN model with the hyperbolic tangent(tanh)function outperforms the CNN models with the other activation functions(the rectified linear unit function,the linear function and the Sigmoid function).Secondly,we compared the performance among the CNN models using different optimization methods.The CNN model with the stochastic gradient descent optimization and the initial learning rate?=?0.0005 achieves satisfactory performance compared to the other CNN models.Finally,we compared the performance of the CNN model,the model based on support vector regression(SVR)and the model based on partial least square regression(PLSR).The results exhibit the CNN model is superior to the SVR model and the PLSR model for all oxides.Based on the above analysis,we conclude the CNN regression model can effectively improve the prediction accuracy of LIBS.
基金Project supported by the National Key R&D Program of China(Grant No.2017YFB0405202)the National Natural Science Foundation of China(Grant Nos.61690221,91850209,and 11774277)。
文摘A convolutional neural network is employed to retrieve the time-domain envelop and phase of few-cycle femtosecond pulses from transient-grating frequency-resolved optical gating(TG-FROG) traces.We use theoretically generated TGFROG traces to complete supervised trainings of the convolutional neural networks,then use similarly generated traces not included in the training dataset to test how well the networks are trained.Accurate retrieval of such traces by the neural network is realized.In our case,we find that networks with exponential linear unit(ELU) activation function perform better than those with leaky rectified linear unit(LRELU) and scaled exponential linear unit(SELU).Finally,the issues that need to be addressed for the retrieval of experimental data by this method are discussed.
基金This work is supported by The National Key Research and Development Program of China(2018YFB1800202,2016YFB1000302,SQ2019ZD090149,2018YFB0204301).
文摘With the continuous development of e-commerce,consumers show increasing interest in posting comments on consumption experience and quality of commodities.Meanwhile,people make purchasing decisions relying on other comments much more than ever before.So the reliability of commodity comments has a significant impact on ensuring consumers’equity and building a fair internet-trade-environment.However,some unscrupulous online-sellers write fake praiseful reviews for themselves and malicious comments for their business counterparts to maximize their profits.Those improper ways of self-profiting have severely ruined the entire online shopping industry.Aiming to detect and prevent these deceptive comments effectively,we construct a model of Multi-Filters Convolutional Neural Network(MFCNN)for opinion spam detection.MFCNN is designed with a fixed-length sequence input and an improved activation function to avoid the gradient vanishing problem in spam opinion detection.Moreover,convolution filters with different widths are used in MFCNN to represent the sentences and documents.Our experimental results show that MFCNN outperforms current state-of-the-art methods on standard spam detection benchmarks.
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.
文摘Traditional methods for judging the degree of insufficient fluid supply in oil wells have low efficiency and limited accuracy. To address this problem, a method for intelligently identifying the degree of insufficient fluid supply in oil wells based on convolutional neural networks is proposed in this paper. Firstly, 5000 indicator diagrams with insufficient liquid supply were collected from the oilfield site, and a sample set was established after preprocessing;then based on the AlexNet model, combined with the characteristics of the indicator diagram, a convolutional neural network model including 4 layers of convolutional layers, 3 layers of down-pooling layers and 2 layers of fully connected layers is established. The backpropagation, ReLu activation function and dropout regularization method are used to complete the training of the convolutional neural network;finally, the performance of the convolutional neural network under different iteration times and network structure is compared, and the super parameter optimization of the model is completed. It has laid a good foundation for realizing the self-adaptive and intelligent matching of oil well production parameters and formation fluid supply conditions. It has certain application prospects. The results show that the accuracy of training and verification of the method exceeds 98%, which can meet the actual application requirements on site.
基金This work was supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.2022R1F1A1062953).
文摘Recently,deep learning has achieved remarkable results in fields that require human cognitive ability,learning ability,and reasoning ability.Activation functions are very important because they provide the ability of artificial neural networks to learn complex patterns through nonlinearity.Various activation functions are being studied to solve problems such as vanishing gradients and dying nodes that may occur in the deep learning process.However,it takes a lot of time and effort for researchers to use the existing activation function in their research.Therefore,in this paper,we propose a universal activation function(UA)so that researchers can easily create and apply various activation functions and improve the performance of neural networks.UA can generate new types of activation functions as well as functions like traditional activation functions by properly adjusting three hyperparameters.The famous Convolutional Neural Network(CNN)and benchmark datasetwere used to evaluate the experimental performance of the UA proposed in this study.We compared the performance of the artificial neural network to which the traditional activation function is applied and the artificial neural network to which theUA is applied.In addition,we evaluated the performance of the new activation function generated by adjusting the hyperparameters of theUA.The experimental performance evaluation results showed that the classification performance of CNNs improved by up to 5%through the UA,although most of them showed similar performance to the traditional activation function.
文摘The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two state-of-the-art methods.
基金funded by the Ongoing Research Funding Program(ORF-2025-890),King Saud University,Riyadh,Saudi Arabia.
文摘Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automation has greatly increased across industries worldwide.Real-time detection requires edge devices with limited computational time.This study proposes a novel hybrid deep learning system for human activity recognition(HAR),aiming to enhance the recognition accuracy and reduce the computational time.The proposed system combines a pretrained image classification model with a sequence analysis model.First,the dataset was divided into a training set(70%),validation set(10%),and test set(20%).Second,all the videos were converted into frames and deep-based features were extracted from each frame using convolutional neural networks(CNNs)with a vision transformer.Following that,bidirectional long short-term memory(BiLSTM)-and temporal convolutional network(TCN)-based models were trained using the training set,and their performances were evaluated using the validation set and test set.Four benchmark datasets(UCF11,UCF50,UCF101,and JHMDB)were used to evaluate the performance of the proposed HAR-based system.The experimental results showed that the combination of ConvNeXt and the TCN-based model achieved a recognition accuracy of 97.73%for UCF11,98.81%for UCF50,98.46%for UCF101,and 83.38%for JHMDB,respectively.This represents improvements in the recognition accuracy of 4%,2.67%,3.67%,and 7.08%for the UCF11,UCF50,UCF101,and JHMDB datasets,respectively,over existing models.Moreover,the proposed HAR-based system obtained superior recognition accuracy,shorter computational times,and minimal memory usage compared to the existing models.
