In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intr...In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intra-pulse modulation signal type based on deep residual network.The basic principle of the recognition method is to obtain the transformation relationship between the time and frequency of complex radar intra-pulse modulation signal through short-time Fourier transform(STFT),and then design an appropriate deep residual network to extract the features of the time-frequency map and complete a variety of complex intra-pulse modulation signal type recognition.In addition,in order to improve the generalization ability of the proposed method,label smoothing and L2 regularization are introduced.The simulation results show that the proposed method has a recognition accuracy of more than 95%for complex radar intra-pulse modulation sig-nal types under low SNR(2 dB).展开更多
Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based...Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based on deep learning to achieve super resolution(SR)by utilizing deep singular-residual neural network(DSRNN)in training phase.Residuals are obtained from the difference between HR and LR images to generate LR-residual example pairs.Singular value decomposition(SVD)is applied to each LR-residual image pair to decompose into subbands of low and high frequency components.Later,DSRNN is trained on these subbands through input and output channels by optimizing the weights and biases of the network.With fewer layers in DSRNN,the influence of exploding gradients is reduced.This speeds up the learning process and also improves accuracy by using skip connections.The trained DSRNN parameters yield residuals to recover the HR subbands in the testing phase.Experimental analysis shows that the proposed method results in superior performance to existingmethods in terms of subjective quality.Extensive testing results on popular benchmark datasets such as set5,set14,and urban100 for a scaling factor of 4 show the effectiveness of the proposed method across different qualitative evaluation metrics.展开更多
With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the...With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the peak in intelligent imaging techniques.However,the presence of noise images degrades both the diagnosis and clinical treatment processes.The existing intelligent meth-ods suffer from the deficiency in handling the diverse range of noise in the ver-satile medical images.This paper proposes a novel deep learning network which learns from the substantial extent of noise in medical data samples to alle-viate this challenge.The proposed deep learning architecture exploits the advan-tages of the capsule network,which is used to extract correlation features and combine them with redefined residual features.Additionally,thefinal stage of dense learning is replaced with powerful extreme learning machines to achieve a better diagnosis rate,even for noisy and complex images.Extensive experimen-tation has been conducted using different medical images.Various performances such as Peak-Signal-To-Noise Ratio(PSNR)and Structural-Similarity-Index-Metrics(SSIM)are compared with the existing deep learning architectures.Addi-tionally,a comprehensive analysis of individual algorithms is analyzed.The experimental results prove that the proposed model has outperformed the other existing algorithms by a substantial margin and proved its supremacy over the other learning models.展开更多
Recognition of human activity is one of the most exciting aspects of time-series classification,with substantial practical and theoretical impli-cations.Recent evidence indicates that activity recognition from wearabl...Recognition of human activity is one of the most exciting aspects of time-series classification,with substantial practical and theoretical impli-cations.Recent evidence indicates that activity recognition from wearable sensors is an effective technique for tracking elderly adults and children in indoor and outdoor environments.Consequently,researchers have demon-strated considerable passion for developing cutting-edge deep learning sys-tems capable of exploiting unprocessed sensor data from wearable devices and generating practical decision assistance in many contexts.This study provides a deep learning-based approach for recognizing indoor and outdoor movement utilizing an enhanced deep pyramidal residual model called Sen-PyramidNet and motion information from wearable sensors(accelerometer and gyroscope).The suggested technique develops a residual unit based on a deep pyramidal residual network and introduces the concept of a pyramidal residual unit to increase detection capability.The proposed deep learning-based model was assessed using the publicly available 19Nonsens dataset,which gathered motion signals from various indoor and outdoor activities,including practicing various body parts.The experimental findings demon-strate that the proposed approach can efficiently reuse characteristics and has achieved an identification accuracy of 96.37%for indoor and 97.25%for outdoor activity.Moreover,comparison experiments demonstrate that the SenPyramidNet surpasses other cutting-edge deep learning models in terms of accuracy and F1-score.Furthermore,this study explores the influence of several wearable sensors on indoor and outdoor action recognition ability.展开更多
Falls are the contributing factor to both fatal and nonfatal injuries in the elderly.Therefore,pre-impact fall detection,which identifies a fall before the body collides with the floor,would be essential.Recently,rese...Falls are the contributing factor to both fatal and nonfatal injuries in the elderly.Therefore,pre-impact fall detection,which identifies a fall before the body collides with the floor,would be essential.Recently,researchers have turned their attention from post-impact fall detection to pre-impact fall detection.Pre-impact fall detection solutions typically use either a threshold-based or machine learning-based approach,although the threshold value would be difficult to accu-rately determine in threshold-based methods.Moreover,while additional features could sometimes assist in categorizing falls and non-falls more precisely,the esti-mated determination of the significant features would be too time-intensive,thus using a significant portion of the algorithm’s operating time.In this work,we developed a deep residual network with aggregation transformation called FDSNeXt for a pre-impact fall detection approach employing wearable inertial sensors.The proposed network was introduced to address the limitations of fea-ture extraction,threshold definition,and algorithm complexity.After training on a large-scale motion dataset,the KFall dataset,and straightforward evaluation with standard metrics,the proposed approach identified pre-impact and impact falls with high accuracy of 91.87 and 92.52%,respectively.In addition,we have inves-tigated fall detection’s performances of three state-of-the-art deep learning models such as a convolutional neural network(CNN),a long short-term memory neural network(LSTM),and a hybrid model(CNN-LSTM).The experimental results showed that the proposed FDSNeXt model outperformed these deep learning models(CNN,LSTM,and CNN-LSTM)with significant improvements.展开更多
With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the...With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the peak in intelligent imaging techniques.However,the presence of noise images degrades both the diagnosis and clinical treatment processes.The existing intelligent meth-ods suffer from the deficiency in handling the diverse range of noise in the ver-satile medical images.This paper proposes a novel deep learning network which learns from the substantial extent of noise in medical data samples to alle-viate this challenge.The proposed deep learning architecture exploits the advan-tages of the capsule network,which is used to extract correlation features and combine them with redefined residual features.Additionally,the final stage of dense learning is replaced with powerful extreme learning machines to achieve a better diagnosis rate,even for noisy and complex images.Extensive experimen-tation has been conducted using different medical images.Various performances such as Peak-Signal-To-Noise Ratio(PSNR)and Structural-Similarity-Index-Metrics(SSIM)are compared with the existing deep learning architectures.Addi-tionally,a comprehensive analysis of individual algorithms is analyzed.The experimental results prove that the proposed model has outperformed the other existing algorithms by a substantial margin and proved its supremacy over the other learning models.展开更多
Rockburst is a phenomenon in which free surfaces are formed during excavation,which subsequently causes the sudden release of energy in the construction of mines and tunnels.Light rockburst only peels off rock slices ...Rockburst is a phenomenon in which free surfaces are formed during excavation,which subsequently causes the sudden release of energy in the construction of mines and tunnels.Light rockburst only peels off rock slices without ejection,while severe rockburst causes casualties and property loss.The frequency and degree of rockburst damage increases with the excavation depth.Moreover,rockburst is the leading engineering geological hazard in the excavation process,and thus the prediction of its intensity grade is of great significance to the development of geotechnical engineering.Therefore,the prediction of rockburst intensity grade is one problem that needs to be solved urgently.By comprehensively considering the occurrence mechanism of rockburst,this paper selects the stress index(σθ/σc),brittleness index(σ_(c)/σ_(t)),and rock elastic energy index(Wet)as the rockburst evaluation indexes through the Spearman coefficient method.This overcomes the low accuracy problem of a single evaluation index prediction method.Following this,the BGD-MSR-DNN rockburst intensity grade prediction model based on batch gradient descent and a multi-scale residual deep neural network is proposed.The batch gradient descent(BGD)module is used to replace the gradient descent algorithm,which effectively improves the efficiency of the network and reduces the model training time.Moreover,the multi-scale residual(MSR)module solves the problem of network degradation when there are too many hidden layers of the deep neural network(DNN),thus improving the model prediction accuracy.The experimental results reveal the BGDMSR-DNN model accuracy to reach 97.1%,outperforming other comparable models.Finally,actual projects such as Qinling Tunnel and Daxiangling Tunnel,reached an accuracy of 100%.The model can be applied in mines and tunnel engineering to realize the accurate and rapid prediction of rockburst intensity grade.展开更多
With the rapid development of deep learning methods, the data-driven approach has shown powerful advantages over the model-driven one. In this paper, we propose an end-to-end autoencoder communication system based on ...With the rapid development of deep learning methods, the data-driven approach has shown powerful advantages over the model-driven one. In this paper, we propose an end-to-end autoencoder communication system based on Deep Residual Shrinkage Networks (DRSNs), where neural networks (DNNs) are used to implement the coding, decoding, modulation and demodulation functions of the communication system. Our proposed autoencoder communication system can better reduce the signal noise by adding an “attention mechanism” and “soft thresholding” modules and has better performance at various signal-to-noise ratios (SNR). Also, we have shown through comparative experiments that the system can operate at moderate block lengths and support different throughputs. It has been shown to work efficiently in the AWGN channel. Simulation results show that our model has a higher Bit-Error-Rate (BER) gain and greatly improved decoding performance compared to conventional modulation and classical autoencoder systems at various signal-to-noise ratios.展开更多
Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face ...Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face great challenges in practical applications due to high computational complexity and dependence on ideal assumptions.This paper presents an effective DOA estimation approach based on a deep residual network(DRN)for the underdetermined case.We first extract an input feature from a new matrix calculated by stacking several covariance matrices corresponding to different time delays.We then provide the input feature to the trained DRN to construct the super resolution spectrum.The DRN learns the mapping relationship between the input feature and the spatial spectrum by training.The proposed approach is superior to existing model-based estimation methods in terms of calculation efficiency,independence of source sparseness and adaptive capacity to non-ideal conditions(e.g.,low signal to noise ratio,short bit sequence).Simulations demonstrate the validity and strong performance of the proposed algorithm on both overdetermined and underdetermined cases.展开更多
Specific emitter identification can distin-guish individual transmitters by analyzing received signals and extracting inherent features of hard-ware circuits.Feature extraction is a key part of traditional machine lea...Specific emitter identification can distin-guish individual transmitters by analyzing received signals and extracting inherent features of hard-ware circuits.Feature extraction is a key part of traditional machine learning-based methods,but manual extrac-tion is generally limited by prior professional knowl-edge.At the same time,it has been noted that the per-formance of most specific emitter identification meth-ods degrades in the low signal-to-noise ratio(SNR)environments.The deep residual shrinkage network(DRSN)is proposed for specific emitter identification,particularly in the low SNRs.The soft threshold can preserve more key features for the improvement of performance,and an identity shortcut can speed up the training process.We collect signals via the receiver to create a dataset in the actual environments.The DRSN is trained to automatically extract features and imple-ment the classification of transmitters.Experimental results show that DRSN obtains the best accuracy un-der different SNRs and has less running time,which demonstrates the effectiveness of DRSN in identify-ing specific emitters.展开更多
In computer vision,object recognition and image categorization have proven to be difficult challenges.They have,nevertheless,generated responses to a wide range of difficult issues from a variety of fields.Convolution...In computer vision,object recognition and image categorization have proven to be difficult challenges.They have,nevertheless,generated responses to a wide range of difficult issues from a variety of fields.Convolution Neural Networks(CNNs)have recently been identified as the most widely proposed deep learning(DL)algorithms in the literature.CNNs have unquestionably delivered cutting-edge achievements,particularly in the areas of image classification,speech recognition,and video processing.However,it has been noticed that the CNN-training assignment demands a large amount of data,which is in low supply,especially in the medical industry,and as a result,the training process takes longer.In this paper,we describe an attentionaware CNN architecture for classifying chest X-ray images to diagnose Pneumonia in order to address the aforementioned difficulties.AttentionModules provide attention-aware properties to the Attention Network.The attentionaware features of various modules alter as the layers become deeper.Using a bottom-up top-down feedforward structure,the feedforward and feedback attention processes are integrated into a single feedforward process inside each attention module.In the present work,a deep neural network(DNN)is combined with an attention mechanism to test the prediction of Pneumonia disease using chest X-ray pictures.To produce attention-aware features,the suggested networkwas built by merging channel and spatial attentionmodules in DNN architecture.With this network,we worked on a publicly available Kaggle chest X-ray dataset.Extensive testing was carried out to validate the suggested model.In the experimental results,we attained an accuracy of 95.47%and an F-score of 0.92,indicating that the suggested model outperformed against the baseline models.展开更多
The accuracy offingerprint recognition model is extremely important due to its usage in forensic and securityfields.Anyfingerprint recognition system has particular network architecture whereas many other networks achiev...The accuracy offingerprint recognition model is extremely important due to its usage in forensic and securityfields.Anyfingerprint recognition system has particular network architecture whereas many other networks achieve higher accuracy.To solve this problem in a unified model,this paper proposes a model that can automatically specify itself.So,it is called an automatic deep neural net-work(ADNN).Our algorithm can specify the appropriate architecture of the neur-al network used and some significant parameters of this network.These parameters are the number offilters,epochs,and iterations.It guarantees the high-est accuracy by updating itself until achieving 99%accuracy then it stops and out-puts the result.Moreover,this paper proposes an end-to-end methodology for recognizing a person’s identity from the inputfingerprint image based on a resi-dual convolutional neural network.It is a complete system and is fully automated whether in the features extraction stage or the classification stage.Our goal is to automate thisfingerprint recognition system because the more automatic the sys-tem is,the more time and effort it saves.Our model also allows users to react by inputting the initial values of these parameters.Then,the model updates itself until itfinds the optimal values for the parameters and achieves the best accuracy.Another advantage of our algorithm is that it can recognize people from their thumb and otherfingers and its ability to recognize distorted samples.Our algo-rithm achieved 99.75%accuracy on the publicfingerprint dataset(SOCOFing).This is the best accuracy compared with other models.展开更多
High frequency(HF) communication is widely spread due to some merits like easy deployment and wide communication coverage. Spectrum prediction is a promising technique to facilitate the working frequency selection and...High frequency(HF) communication is widely spread due to some merits like easy deployment and wide communication coverage. Spectrum prediction is a promising technique to facilitate the working frequency selection and enhance the function of automatic link establishment. Most of the existing spectrum prediction algorithms focus on predicting spectrum values in a slot-by-slot manner and therefore are lack of timeliness. Deep learning based spectrum prediction is developed in this paper by simultaneously predicting multi-slot ahead states of multiple spectrum points within a period of time. Specifically, we first employ supervised learning and construct samples depending on longterm and short-term HF spectrum data. Then, advanced residual units are introduced to build multiple residual network modules to respectively capture characteristics in these data with diverse time scales. Further, convolution neural network fuses the outputs of residual network modules above for temporal-spectral prediction, which is combined with residual network modules to construct the deep temporal-spectral residual network. Experiments have demonstrated that the approach proposed in this paper has a significant advantage over the benchmark schemes.展开更多
alient object detection aims at identifying the visually interesting object regions that are consistent with human perception. Multispectral remote sensing images provide rich radiometric information in revealing the ...alient object detection aims at identifying the visually interesting object regions that are consistent with human perception. Multispectral remote sensing images provide rich radiometric information in revealing the physical properties of the observed objects, which leads to great potential to perform salient object detection for remote sensing images. Conventional salient object detection methods often employ handcrafted features to predict saliency by evaluating the pixel-wise or superpixel-wise contrast. With the recent use of deep learning framework, in particular, fully convolutional neural networks, there has been profound progress in visual saliency detection. However, this success has not been extended to multispectral remote sensing images, and existing multispectral salient object detection methods are still mainly based on handcrafted features, essentially due to the difficulties in image acquisition and labeling. In this paper, we propose a novel deep residual network based on a top-down model, which is trained in an end-to-end manner to tackle the above issues in multispectral salient object detection. Our model effectively exploits the saliency cues at different levels of the deep residual network. To overcome the limited availability of remote sensing images in training of our deep residual network, we also introduce a new spectral image reconstruction model that can generate multispectral images from RGB images. Our extensive experimental results using both multispectral and RGB salient object detection datasets demonstrate a significant performance improvement of more than 10% improvement compared with the state-of-the-art methods.展开更多
Encrypted traffic identification pertains to the precise acquisition and categorization of data from traffic datasets containing imbalanced and obscured content.The extraction of encrypted traffic attributes and their...Encrypted traffic identification pertains to the precise acquisition and categorization of data from traffic datasets containing imbalanced and obscured content.The extraction of encrypted traffic attributes and their subsequent identification presents a formidable challenge.The existing models have predominantly relied on direct extraction of encrypted traffic data from imbalanced datasets,with the dataset’s imbalance significantly affecting the model’s performance.In the present study,a new model,referred to as UD-VLD(Unbalanced Dataset-VAE-LSTM-DRN),was proposed to address above problem.The proposed model is an encrypted traffic identification model for handling unbalanced datasets.The encoder of the variational autoencoder(VAE)is combined with the decoder and Long-short term Memory(LSTM)in UD-VLD model to realize the data enhancement processing of the original unbalanced datasets.The enhanced data is processed by transforming the deep residual network(DRN)to address neural network gradient-related issues.Subsequently,the data is classified and recognized.The UD-VLD model integrates the related techniques of deep learning into the encrypted traffic recognition technique,thereby solving the processing problem for unbalanced datasets.The UD-VLD model was tested using the publicly available Tor dataset and VPN dataset.The UD-VLD model is evaluated against other comparative models in terms of accuracy,loss rate,precision,recall,F1-score,total time,and ROC curve.The results reveal that the UD-VLD model exhibits better performance in both binary and multi classification,being higher than other encrypted traffic recognition models that exist for unbalanced datasets.Furthermore,the evaluation performance indicates that the UD-VLD model effectivelymitigates the impact of unbalanced data on traffic classification.and can serve as a novel solution for encrypted traffic identification.展开更多
In this paper,we propose an improved deep residual network model to recognize human actions.Action data is composed of channel state information signals,which are continuous fine-grained signals.We replaced the tradit...In this paper,we propose an improved deep residual network model to recognize human actions.Action data is composed of channel state information signals,which are continuous fine-grained signals.We replaced the traditional identity connection with the shrinking thresholdmodule.Themodule automatically adjusts the threshold of the action data signal,and filters out signals that are not related to the principal components.We use the attention mechanism to improve the memory of the network model to the action signal,so as to better recognize the action.To verify the validity of the experiment more accurately,we collected action data in two different environments.The experimental results show that the improved network model is much better than the traditional network in recognition.The accuracy of recognition in complex places can reach 92.85%,among which the recognition rate of raising hands is up to 96%.We combine the improved residual deep network model with channel state information action data,and prove the effectiveness of our model for classification through experimental data.展开更多
In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and un...In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and unified mapping of different modes:A Cross-Modal Hashing retrieval algorithm based on Deep Residual Network(CMHR-DRN).The model construction is divided into two stages:The first stage is the feature extraction of different modal data,including the use of Deep Residual Network(DRN)to extract the image features,using the method of combining TF-IDF with the full connection network to extract the text features,and the obtained image and text features used as the input of the second stage.In the second stage,the image and text features are mapped into Hash functions by supervised learning,and the image and text features are mapped to the common binary Hamming space.In the process of mapping,the distance measurement of the original distance measurement and the common feature space are kept unchanged as far as possible to improve the accuracy of Cross-Modal Retrieval.In training the model,adaptive moment estimation(Adam)is used to calculate the adaptive learning rate of each parameter,and the stochastic gradient descent(SGD)is calculated to obtain the minimum loss function.The whole training process is completed on Caffe deep learning framework.Experiments show that the proposed algorithm CMHR-DRN based on Deep Residual Network has better retrieval performance and stronger advantages than other Cross-Modal algorithms CMFH,CMDN and CMSSH.展开更多
Even though much advancements have been achieved with regards to the recognition of handwritten characters,researchers still face difficulties with the handwritten character recognition problem,especially with the adv...Even though much advancements have been achieved with regards to the recognition of handwritten characters,researchers still face difficulties with the handwritten character recognition problem,especially with the advent of new datasets like the Extended Modified National Institute of Standards and Technology dataset(EMNIST).The EMNIST dataset represents a challenge for both machine-learning and deep-learning techniques due to inter-class similarity and intra-class variability.Inter-class similarity exists because of the similarity between the shapes of certain characters in the dataset.The presence of intra-class variability is mainly due to different shapes written by different writers for the same character.In this research,we have optimized a deep residual network to achieve higher accuracy vs.the published state-of-the-art results.This approach is mainly based on the prebuilt deep residual network model ResNet18,whose architecture has been enhanced by using the optimal number of residual blocks and the optimal size of the receptive field of the first convolutional filter,the replacement of the first max-pooling filter by an average pooling filter,and the addition of a drop-out layer before the fully connected layer.A distinctive modification has been introduced by replacing the final addition layer with a depth concatenation layer,which resulted in a novel deep architecture having higher accuracy vs.the pure residual architecture.Moreover,the dataset images’sizes have been adjusted to optimize their visibility in the network.Finally,by tuning the training hyperparameters and using rotation and shear augmentations,the proposed model outperformed the state-of-the-art models by achieving average accuracies of 95.91%and 90.90%for the Letters and Balanced dataset sections,respectively.Furthermore,the average accuracies were improved to 95.9%and 91.06%for the Letters and Balanced sections,respectively,by using a group of 5 instances of the trained models and averaging the output class probabilities.展开更多
The 3D sand printing(3DSP),by binder jetting technology for rapid casting,has a pivotal role in promoting the development of the traditional casting industry as a result of producing high-quality and economical sand m...The 3D sand printing(3DSP),by binder jetting technology for rapid casting,has a pivotal role in promoting the development of the traditional casting industry as a result of producing high-quality and economical sand molds.This work presents an approach for monitoring and analyzing powder sand-bed images to serve as a real-time control system in a 3DSP machine.A deep residual network(ResNet)is used to classify the defects occurring during the powder spreading stage of the process.Firstly,a pre-trained network was applied as the initial parameter;then it was fine-tuned on the labelled defective sample dataset to accomplish the task,which defines the sand-bed defects induced in the 3DSP processing.Furthermore,the recognition and positioning of sand-bed defects were readily achieved by dividing the sand-bed images into blocks.Experiments show that the fine-tuned network has a 98.7%classification accuracy on the validation dataset of sand-bed defects and 95.4%recognition accuracy for the sand-bed images.展开更多
文摘In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),the paper proposes an automatic recog-nition method of complex radar intra-pulse modulation signal type based on deep residual network.The basic principle of the recognition method is to obtain the transformation relationship between the time and frequency of complex radar intra-pulse modulation signal through short-time Fourier transform(STFT),and then design an appropriate deep residual network to extract the features of the time-frequency map and complete a variety of complex intra-pulse modulation signal type recognition.In addition,in order to improve the generalization ability of the proposed method,label smoothing and L2 regularization are introduced.The simulation results show that the proposed method has a recognition accuracy of more than 95%for complex radar intra-pulse modulation sig-nal types under low SNR(2 dB).
文摘Single image super resolution(SISR)techniques produce images of high resolution(HR)as output from input images of low resolution(LR).Motivated by the effectiveness of deep learning methods,we provide a framework based on deep learning to achieve super resolution(SR)by utilizing deep singular-residual neural network(DSRNN)in training phase.Residuals are obtained from the difference between HR and LR images to generate LR-residual example pairs.Singular value decomposition(SVD)is applied to each LR-residual image pair to decompose into subbands of low and high frequency components.Later,DSRNN is trained on these subbands through input and output channels by optimizing the weights and biases of the network.With fewer layers in DSRNN,the influence of exploding gradients is reduced.This speeds up the learning process and also improves accuracy by using skip connections.The trained DSRNN parameters yield residuals to recover the HR subbands in the testing phase.Experimental analysis shows that the proposed method results in superior performance to existingmethods in terms of subjective quality.Extensive testing results on popular benchmark datasets such as set5,set14,and urban100 for a scaling factor of 4 show the effectiveness of the proposed method across different qualitative evaluation metrics.
文摘With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the peak in intelligent imaging techniques.However,the presence of noise images degrades both the diagnosis and clinical treatment processes.The existing intelligent meth-ods suffer from the deficiency in handling the diverse range of noise in the ver-satile medical images.This paper proposes a novel deep learning network which learns from the substantial extent of noise in medical data samples to alle-viate this challenge.The proposed deep learning architecture exploits the advan-tages of the capsule network,which is used to extract correlation features and combine them with redefined residual features.Additionally,thefinal stage of dense learning is replaced with powerful extreme learning machines to achieve a better diagnosis rate,even for noisy and complex images.Extensive experimen-tation has been conducted using different medical images.Various performances such as Peak-Signal-To-Noise Ratio(PSNR)and Structural-Similarity-Index-Metrics(SSIM)are compared with the existing deep learning architectures.Addi-tionally,a comprehensive analysis of individual algorithms is analyzed.The experimental results prove that the proposed model has outperformed the other existing algorithms by a substantial margin and proved its supremacy over the other learning models.
基金supported by the Thailand Science Research and Innovation Fundthe University of Phayao(Grant No.FF66-UoE001)King Mongkut’s University of Technology North Bangkok,Contract No.KMUTNB-66-KNOW-05.
文摘Recognition of human activity is one of the most exciting aspects of time-series classification,with substantial practical and theoretical impli-cations.Recent evidence indicates that activity recognition from wearable sensors is an effective technique for tracking elderly adults and children in indoor and outdoor environments.Consequently,researchers have demon-strated considerable passion for developing cutting-edge deep learning sys-tems capable of exploiting unprocessed sensor data from wearable devices and generating practical decision assistance in many contexts.This study provides a deep learning-based approach for recognizing indoor and outdoor movement utilizing an enhanced deep pyramidal residual model called Sen-PyramidNet and motion information from wearable sensors(accelerometer and gyroscope).The suggested technique develops a residual unit based on a deep pyramidal residual network and introduces the concept of a pyramidal residual unit to increase detection capability.The proposed deep learning-based model was assessed using the publicly available 19Nonsens dataset,which gathered motion signals from various indoor and outdoor activities,including practicing various body parts.The experimental findings demon-strate that the proposed approach can efficiently reuse characteristics and has achieved an identification accuracy of 96.37%for indoor and 97.25%for outdoor activity.Moreover,comparison experiments demonstrate that the SenPyramidNet surpasses other cutting-edge deep learning models in terms of accuracy and F1-score.Furthermore,this study explores the influence of several wearable sensors on indoor and outdoor action recognition ability.
基金This research project was also supported by the Thailand Science Research and Innovation Fundthe University of Phayao(Grant No.FF66-UoE001)King Mongkut’s University of Technology North Bangkok under Contract No.KMUTNB-66-KNOW-05.
文摘Falls are the contributing factor to both fatal and nonfatal injuries in the elderly.Therefore,pre-impact fall detection,which identifies a fall before the body collides with the floor,would be essential.Recently,researchers have turned their attention from post-impact fall detection to pre-impact fall detection.Pre-impact fall detection solutions typically use either a threshold-based or machine learning-based approach,although the threshold value would be difficult to accu-rately determine in threshold-based methods.Moreover,while additional features could sometimes assist in categorizing falls and non-falls more precisely,the esti-mated determination of the significant features would be too time-intensive,thus using a significant portion of the algorithm’s operating time.In this work,we developed a deep residual network with aggregation transformation called FDSNeXt for a pre-impact fall detection approach employing wearable inertial sensors.The proposed network was introduced to address the limitations of fea-ture extraction,threshold definition,and algorithm complexity.After training on a large-scale motion dataset,the KFall dataset,and straightforward evaluation with standard metrics,the proposed approach identified pre-impact and impact falls with high accuracy of 91.87 and 92.52%,respectively.In addition,we have inves-tigated fall detection’s performances of three state-of-the-art deep learning models such as a convolutional neural network(CNN),a long short-term memory neural network(LSTM),and a hybrid model(CNN-LSTM).The experimental results showed that the proposed FDSNeXt model outperformed these deep learning models(CNN,LSTM,and CNN-LSTM)with significant improvements.
文摘With the advent of Machine and Deep Learning algorithms,medical image diagnosis has a new perception of diagnosis and clinical treatment.Regret-tably,medical images are more susceptible to capturing noises despite the peak in intelligent imaging techniques.However,the presence of noise images degrades both the diagnosis and clinical treatment processes.The existing intelligent meth-ods suffer from the deficiency in handling the diverse range of noise in the ver-satile medical images.This paper proposes a novel deep learning network which learns from the substantial extent of noise in medical data samples to alle-viate this challenge.The proposed deep learning architecture exploits the advan-tages of the capsule network,which is used to extract correlation features and combine them with redefined residual features.Additionally,the final stage of dense learning is replaced with powerful extreme learning machines to achieve a better diagnosis rate,even for noisy and complex images.Extensive experimen-tation has been conducted using different medical images.Various performances such as Peak-Signal-To-Noise Ratio(PSNR)and Structural-Similarity-Index-Metrics(SSIM)are compared with the existing deep learning architectures.Addi-tionally,a comprehensive analysis of individual algorithms is analyzed.The experimental results prove that the proposed model has outperformed the other existing algorithms by a substantial margin and proved its supremacy over the other learning models.
基金funded by State Key Laboratory for GeoMechanics and Deep Underground Engineering&Institute for Deep Underground Science and Engineering,Grant Number XD2021021BUCEA Post Graduate Innovation Project under Grant,Grant Number PG2023092.
文摘Rockburst is a phenomenon in which free surfaces are formed during excavation,which subsequently causes the sudden release of energy in the construction of mines and tunnels.Light rockburst only peels off rock slices without ejection,while severe rockburst causes casualties and property loss.The frequency and degree of rockburst damage increases with the excavation depth.Moreover,rockburst is the leading engineering geological hazard in the excavation process,and thus the prediction of its intensity grade is of great significance to the development of geotechnical engineering.Therefore,the prediction of rockburst intensity grade is one problem that needs to be solved urgently.By comprehensively considering the occurrence mechanism of rockburst,this paper selects the stress index(σθ/σc),brittleness index(σ_(c)/σ_(t)),and rock elastic energy index(Wet)as the rockburst evaluation indexes through the Spearman coefficient method.This overcomes the low accuracy problem of a single evaluation index prediction method.Following this,the BGD-MSR-DNN rockburst intensity grade prediction model based on batch gradient descent and a multi-scale residual deep neural network is proposed.The batch gradient descent(BGD)module is used to replace the gradient descent algorithm,which effectively improves the efficiency of the network and reduces the model training time.Moreover,the multi-scale residual(MSR)module solves the problem of network degradation when there are too many hidden layers of the deep neural network(DNN),thus improving the model prediction accuracy.The experimental results reveal the BGDMSR-DNN model accuracy to reach 97.1%,outperforming other comparable models.Finally,actual projects such as Qinling Tunnel and Daxiangling Tunnel,reached an accuracy of 100%.The model can be applied in mines and tunnel engineering to realize the accurate and rapid prediction of rockburst intensity grade.
文摘With the rapid development of deep learning methods, the data-driven approach has shown powerful advantages over the model-driven one. In this paper, we propose an end-to-end autoencoder communication system based on Deep Residual Shrinkage Networks (DRSNs), where neural networks (DNNs) are used to implement the coding, decoding, modulation and demodulation functions of the communication system. Our proposed autoencoder communication system can better reduce the signal noise by adding an “attention mechanism” and “soft thresholding” modules and has better performance at various signal-to-noise ratios (SNR). Also, we have shown through comparative experiments that the system can operate at moderate block lengths and support different throughputs. It has been shown to work efficiently in the AWGN channel. Simulation results show that our model has a higher Bit-Error-Rate (BER) gain and greatly improved decoding performance compared to conventional modulation and classical autoencoder systems at various signal-to-noise ratios.
基金supported by the Fundamental Research Funds for the Central Universities (No.2022JCCXMT01)the National College Students’Innovation and Entrepreneurship Training Program Automatic Recognition of Earthquake Faults Based on Convolutional Neural Networks (No.20220236)the Open Fund of State Key Laboratory for Fine Exploration and Intelligent Development of Coal Resources (No.SKLCRSM22DC02).
基金supported by the Program for Innovative Research Groups of the Hunan Provincial Natural Science Foundation of China(2019JJ10004)。
文摘Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face great challenges in practical applications due to high computational complexity and dependence on ideal assumptions.This paper presents an effective DOA estimation approach based on a deep residual network(DRN)for the underdetermined case.We first extract an input feature from a new matrix calculated by stacking several covariance matrices corresponding to different time delays.We then provide the input feature to the trained DRN to construct the super resolution spectrum.The DRN learns the mapping relationship between the input feature and the spatial spectrum by training.The proposed approach is superior to existing model-based estimation methods in terms of calculation efficiency,independence of source sparseness and adaptive capacity to non-ideal conditions(e.g.,low signal to noise ratio,short bit sequence).Simulations demonstrate the validity and strong performance of the proposed algorithm on both overdetermined and underdetermined cases.
基金the National Natural Science Foundation of China(No.U20B2038,No.61871398,NO.61901520 and No.61931011)the Natural Science Foundation for Distinguished Young Scholars of Jiangsu Province(No.BK20190030)the National Key R&D Program of China under Grant 2018YFB1801103.
文摘Specific emitter identification can distin-guish individual transmitters by analyzing received signals and extracting inherent features of hard-ware circuits.Feature extraction is a key part of traditional machine learning-based methods,but manual extrac-tion is generally limited by prior professional knowl-edge.At the same time,it has been noted that the per-formance of most specific emitter identification meth-ods degrades in the low signal-to-noise ratio(SNR)environments.The deep residual shrinkage network(DRSN)is proposed for specific emitter identification,particularly in the low SNRs.The soft threshold can preserve more key features for the improvement of performance,and an identity shortcut can speed up the training process.We collect signals via the receiver to create a dataset in the actual environments.The DRSN is trained to automatically extract features and imple-ment the classification of transmitters.Experimental results show that DRSN obtains the best accuracy un-der different SNRs and has less running time,which demonstrates the effectiveness of DRSN in identify-ing specific emitters.
文摘In computer vision,object recognition and image categorization have proven to be difficult challenges.They have,nevertheless,generated responses to a wide range of difficult issues from a variety of fields.Convolution Neural Networks(CNNs)have recently been identified as the most widely proposed deep learning(DL)algorithms in the literature.CNNs have unquestionably delivered cutting-edge achievements,particularly in the areas of image classification,speech recognition,and video processing.However,it has been noticed that the CNN-training assignment demands a large amount of data,which is in low supply,especially in the medical industry,and as a result,the training process takes longer.In this paper,we describe an attentionaware CNN architecture for classifying chest X-ray images to diagnose Pneumonia in order to address the aforementioned difficulties.AttentionModules provide attention-aware properties to the Attention Network.The attentionaware features of various modules alter as the layers become deeper.Using a bottom-up top-down feedforward structure,the feedforward and feedback attention processes are integrated into a single feedforward process inside each attention module.In the present work,a deep neural network(DNN)is combined with an attention mechanism to test the prediction of Pneumonia disease using chest X-ray pictures.To produce attention-aware features,the suggested networkwas built by merging channel and spatial attentionmodules in DNN architecture.With this network,we worked on a publicly available Kaggle chest X-ray dataset.Extensive testing was carried out to validate the suggested model.In the experimental results,we attained an accuracy of 95.47%and an F-score of 0.92,indicating that the suggested model outperformed against the baseline models.
文摘The accuracy offingerprint recognition model is extremely important due to its usage in forensic and securityfields.Anyfingerprint recognition system has particular network architecture whereas many other networks achieve higher accuracy.To solve this problem in a unified model,this paper proposes a model that can automatically specify itself.So,it is called an automatic deep neural net-work(ADNN).Our algorithm can specify the appropriate architecture of the neur-al network used and some significant parameters of this network.These parameters are the number offilters,epochs,and iterations.It guarantees the high-est accuracy by updating itself until achieving 99%accuracy then it stops and out-puts the result.Moreover,this paper proposes an end-to-end methodology for recognizing a person’s identity from the inputfingerprint image based on a resi-dual convolutional neural network.It is a complete system and is fully automated whether in the features extraction stage or the classification stage.Our goal is to automate thisfingerprint recognition system because the more automatic the sys-tem is,the more time and effort it saves.Our model also allows users to react by inputting the initial values of these parameters.Then,the model updates itself until itfinds the optimal values for the parameters and achieves the best accuracy.Another advantage of our algorithm is that it can recognize people from their thumb and otherfingers and its ability to recognize distorted samples.Our algo-rithm achieved 99.75%accuracy on the publicfingerprint dataset(SOCOFing).This is the best accuracy compared with other models.
基金supported in part by the National Natural Science Foundation of China (Grants No. 61501510 and No. 61631020)Natural Science Foundation of Jiangsu Province (Grant No. BK20150717)+2 种基金China Postdoctoral Science Foundation Funded Project (Grant No. 2016M590398 and No.2018T110426)Jiangsu Planned Projects for Postdoctoral Research Funds (Grant No. 1501009A)Natural Science Foundation for Distinguished Young Scholars of Jiangsu Province (Grant No. BK20160034)
文摘High frequency(HF) communication is widely spread due to some merits like easy deployment and wide communication coverage. Spectrum prediction is a promising technique to facilitate the working frequency selection and enhance the function of automatic link establishment. Most of the existing spectrum prediction algorithms focus on predicting spectrum values in a slot-by-slot manner and therefore are lack of timeliness. Deep learning based spectrum prediction is developed in this paper by simultaneously predicting multi-slot ahead states of multiple spectrum points within a period of time. Specifically, we first employ supervised learning and construct samples depending on longterm and short-term HF spectrum data. Then, advanced residual units are introduced to build multiple residual network modules to respectively capture characteristics in these data with diverse time scales. Further, convolution neural network fuses the outputs of residual network modules above for temporal-spectral prediction, which is combined with residual network modules to construct the deep temporal-spectral residual network. Experiments have demonstrated that the approach proposed in this paper has a significant advantage over the benchmark schemes.
基金National 1000 Young Talents Plan of ChinaNational Natural Science Foundation of China(61420106007,61671387,61871325)DECRA of Australica Resenrch Council (DE140100180).
文摘alient object detection aims at identifying the visually interesting object regions that are consistent with human perception. Multispectral remote sensing images provide rich radiometric information in revealing the physical properties of the observed objects, which leads to great potential to perform salient object detection for remote sensing images. Conventional salient object detection methods often employ handcrafted features to predict saliency by evaluating the pixel-wise or superpixel-wise contrast. With the recent use of deep learning framework, in particular, fully convolutional neural networks, there has been profound progress in visual saliency detection. However, this success has not been extended to multispectral remote sensing images, and existing multispectral salient object detection methods are still mainly based on handcrafted features, essentially due to the difficulties in image acquisition and labeling. In this paper, we propose a novel deep residual network based on a top-down model, which is trained in an end-to-end manner to tackle the above issues in multispectral salient object detection. Our model effectively exploits the saliency cues at different levels of the deep residual network. To overcome the limited availability of remote sensing images in training of our deep residual network, we also introduce a new spectral image reconstruction model that can generate multispectral images from RGB images. Our extensive experimental results using both multispectral and RGB salient object detection datasets demonstrate a significant performance improvement of more than 10% improvement compared with the state-of-the-art methods.
基金supported by the Fundamental Research Funds for Higher Education Institutions of Heilongjiang Province(145209126)the Heilongjiang Province Higher Education Teaching Reform Project under Grant No.SJGY20200770.
文摘Encrypted traffic identification pertains to the precise acquisition and categorization of data from traffic datasets containing imbalanced and obscured content.The extraction of encrypted traffic attributes and their subsequent identification presents a formidable challenge.The existing models have predominantly relied on direct extraction of encrypted traffic data from imbalanced datasets,with the dataset’s imbalance significantly affecting the model’s performance.In the present study,a new model,referred to as UD-VLD(Unbalanced Dataset-VAE-LSTM-DRN),was proposed to address above problem.The proposed model is an encrypted traffic identification model for handling unbalanced datasets.The encoder of the variational autoencoder(VAE)is combined with the decoder and Long-short term Memory(LSTM)in UD-VLD model to realize the data enhancement processing of the original unbalanced datasets.The enhanced data is processed by transforming the deep residual network(DRN)to address neural network gradient-related issues.Subsequently,the data is classified and recognized.The UD-VLD model integrates the related techniques of deep learning into the encrypted traffic recognition technique,thereby solving the processing problem for unbalanced datasets.The UD-VLD model was tested using the publicly available Tor dataset and VPN dataset.The UD-VLD model is evaluated against other comparative models in terms of accuracy,loss rate,precision,recall,F1-score,total time,and ROC curve.The results reveal that the UD-VLD model exhibits better performance in both binary and multi classification,being higher than other encrypted traffic recognition models that exist for unbalanced datasets.Furthermore,the evaluation performance indicates that the UD-VLD model effectivelymitigates the impact of unbalanced data on traffic classification.and can serve as a novel solution for encrypted traffic identification.
基金This work was supported by Innovation Capability Support Program of Shaanxi(Program No.2018TD-016)Key Research and Development Program of Shaanxi(Program No.2019ZDLSF02-09-02).
文摘In this paper,we propose an improved deep residual network model to recognize human actions.Action data is composed of channel state information signals,which are continuous fine-grained signals.We replaced the traditional identity connection with the shrinking thresholdmodule.Themodule automatically adjusts the threshold of the action data signal,and filters out signals that are not related to the principal components.We use the attention mechanism to improve the memory of the network model to the action signal,so as to better recognize the action.To verify the validity of the experiment more accurately,we collected action data in two different environments.The experimental results show that the improved network model is much better than the traditional network in recognition.The accuracy of recognition in complex places can reach 92.85%,among which the recognition rate of raising hands is up to 96%.We combine the improved residual deep network model with channel state information action data,and prove the effectiveness of our model for classification through experimental data.
文摘In the era of big data rich inWe Media,the single mode retrieval system has been unable to meet people’s demand for information retrieval.This paper proposes a new solution to the problem of feature extraction and unified mapping of different modes:A Cross-Modal Hashing retrieval algorithm based on Deep Residual Network(CMHR-DRN).The model construction is divided into two stages:The first stage is the feature extraction of different modal data,including the use of Deep Residual Network(DRN)to extract the image features,using the method of combining TF-IDF with the full connection network to extract the text features,and the obtained image and text features used as the input of the second stage.In the second stage,the image and text features are mapped into Hash functions by supervised learning,and the image and text features are mapped to the common binary Hamming space.In the process of mapping,the distance measurement of the original distance measurement and the common feature space are kept unchanged as far as possible to improve the accuracy of Cross-Modal Retrieval.In training the model,adaptive moment estimation(Adam)is used to calculate the adaptive learning rate of each parameter,and the stochastic gradient descent(SGD)is calculated to obtain the minimum loss function.The whole training process is completed on Caffe deep learning framework.Experiments show that the proposed algorithm CMHR-DRN based on Deep Residual Network has better retrieval performance and stronger advantages than other Cross-Modal algorithms CMFH,CMDN and CMSSH.
文摘Even though much advancements have been achieved with regards to the recognition of handwritten characters,researchers still face difficulties with the handwritten character recognition problem,especially with the advent of new datasets like the Extended Modified National Institute of Standards and Technology dataset(EMNIST).The EMNIST dataset represents a challenge for both machine-learning and deep-learning techniques due to inter-class similarity and intra-class variability.Inter-class similarity exists because of the similarity between the shapes of certain characters in the dataset.The presence of intra-class variability is mainly due to different shapes written by different writers for the same character.In this research,we have optimized a deep residual network to achieve higher accuracy vs.the published state-of-the-art results.This approach is mainly based on the prebuilt deep residual network model ResNet18,whose architecture has been enhanced by using the optimal number of residual blocks and the optimal size of the receptive field of the first convolutional filter,the replacement of the first max-pooling filter by an average pooling filter,and the addition of a drop-out layer before the fully connected layer.A distinctive modification has been introduced by replacing the final addition layer with a depth concatenation layer,which resulted in a novel deep architecture having higher accuracy vs.the pure residual architecture.Moreover,the dataset images’sizes have been adjusted to optimize their visibility in the network.Finally,by tuning the training hyperparameters and using rotation and shear augmentations,the proposed model outperformed the state-of-the-art models by achieving average accuracies of 95.91%and 90.90%for the Letters and Balanced dataset sections,respectively.Furthermore,the average accuracies were improved to 95.9%and 91.06%for the Letters and Balanced sections,respectively,by using a group of 5 instances of the trained models and averaging the output class probabilities.
文摘The 3D sand printing(3DSP),by binder jetting technology for rapid casting,has a pivotal role in promoting the development of the traditional casting industry as a result of producing high-quality and economical sand molds.This work presents an approach for monitoring and analyzing powder sand-bed images to serve as a real-time control system in a 3DSP machine.A deep residual network(ResNet)is used to classify the defects occurring during the powder spreading stage of the process.Firstly,a pre-trained network was applied as the initial parameter;then it was fine-tuned on the labelled defective sample dataset to accomplish the task,which defines the sand-bed defects induced in the 3DSP processing.Furthermore,the recognition and positioning of sand-bed defects were readily achieved by dividing the sand-bed images into blocks.Experiments show that the fine-tuned network has a 98.7%classification accuracy on the validation dataset of sand-bed defects and 95.4%recognition accuracy for the sand-bed images.