Brain tumors pose significant diagnostic challenges due to their diverse types and complex anatomical locations.Due to the increase in precision image-based diagnostic tools,driven by advancements in artificial intell...Brain tumors pose significant diagnostic challenges due to their diverse types and complex anatomical locations.Due to the increase in precision image-based diagnostic tools,driven by advancements in artificial intelligence(AI)and deep learning,there has been potential to improve diagnostic accuracy,especially with Magnetic Resonance Imaging(MRI).However,traditional state-of-the-art models lack the sensitivity essential for reliable tumor identification and segmentation.Thus,our research aims to enhance brain tumor diagnosis in MRI by proposing an advanced model.The proposed model incorporates dilated convolutions to optimize the brain tumor segmentation and classification.The proposed model is first trained and later evaluated using the BraTS 2020 dataset.In our proposed model preprocessing consists of normalization,noise reduction,and data augmentation to improve model robustness.The attention mechanism and dilated convolutions were introduced to increase the model’s focus on critical regions and capture finer spatial details without compromising image resolution.We have performed experimentation to measure efficiency.For this,we have used various metrics including accuracy,sensitivity,and curve(AUC-ROC).The proposed model achieved a high accuracy of 94%,a sensitivity of 93%,a specificity of 92%,and an AUC-ROC of 0.98,outperforming traditional diagnostic models in brain tumor detection.The proposed model accurately identifies tumor regions,while dilated convolutions enhanced the segmentation accuracy,especially for complex tumor structures.The proposed model demonstrates significant potential for clinical application,providing reliable and precise brain tumor detection in MRI.展开更多
The Bernoulli convolution ν λ measure is shown to be absolutely continuous with L 2 density for almost all 12<λ<1,and singular if λ -1 is a Pisot number. It is an open question whether the Pisot typ...The Bernoulli convolution ν λ measure is shown to be absolutely continuous with L 2 density for almost all 12<λ<1,and singular if λ -1 is a Pisot number. It is an open question whether the Pisot type Bernoulli convolutions are the only singular ones. In this paper,we construct a family of non-Pisot type Bernoulli convolutions ν λ such that their density functions,if they exist,are not L 2. We also construct other Bernolulli convolutions whose density functions,if they exist,behave rather badly.展开更多
Louis Pierre Gratiolet (1815-1865) was one of the first modern anatomists to pay attention to cerebral convolutions. Born in Sainte-Foy-la-Grande (Gironde), he moved to Paris in 1834 to study medicine, as well as comp...Louis Pierre Gratiolet (1815-1865) was one of the first modern anatomists to pay attention to cerebral convolutions. Born in Sainte-Foy-la-Grande (Gironde), he moved to Paris in 1834 to study medicine, as well as comparative anatomy under Henri de Blainville (1777-1850). In 1842, he accepted de Blainville’s offer to become his assistant at the Muséum d’histoire naturelle and progressively abandoned medicine for comparative anatomy. He undertook a detailed study of brains of human and nonhuman primates and soon realized that the organizational pattern of cerebral convolutions was so predictable that it could serve as a criterion to classify primate groups. He noted that only the deepest sulci exist in lower primate forms, while the complexity of cortical folding increases markedly in great apes and humans. Gratiolet provided the first cogent description of the lobular organization of primate cerebral hemispheres. He saw the insula as a central lobe around which revolved the frontal, parietal, temporal (temporo-sphenoidal) and occipital lobes. He correctly identified most gyri and sulci on all brain surfaces, introduced the term “plis de passage” for some interconnecting gyri, and provided the first description of the optic radiations. In the early 1860s, Gratiolet fought a highly publicized battle against Paul Broca (1824-1880) on the relationship between brain and intelligence. Gratiolet agreed that the brain was most likely the seat of intelligence, but he considered human cognition far too subtle to have any direct relationship with brain size. He argued that a detailed study of the human brain architecture would be more profitable than Broca’s vain speculations on the relationship between brain weight and intelligence, which he considered a monolithic entity. Despite remarkable scientific achievements and a unique teaching capacity, Gratiolet was unable to secure any academic position until three years before his sudden death in Paris at age 49.展开更多
Pointwise convolution is usually utilized to expand or squeeze features in modern lightweight deep models.However,it takes up most of the overall computational cost(usually more than 90%).This paper proposes a novel P...Pointwise convolution is usually utilized to expand or squeeze features in modern lightweight deep models.However,it takes up most of the overall computational cost(usually more than 90%).This paper proposes a novel Poker module to expand features by taking advantage of cheap depthwise convolution.As a result,the Poker module can greatly reduce the computational cost,and meanwhile generate a large number of effective features to guarantee the performance.The proposed module is standardized and can be employed wherever the feature expansion is needed.By varying the stride and the number of channels,different kinds of bottlenecks are designed to plug the proposed Poker module into the network.Thus,a lightweight model can be easily assembled.Experiments conducted on benchmarks reveal the effectiveness of our proposed Poker module.And our Poker Net models can reduce the computational cost by 7.1%-15.6%.Poker Net models achieve comparable or even higher recognition accuracy than previous state-of-the-art(SOTA)models on the Image Net ILSVRC2012 classification dataset.Code is available at https://github.com/diaomin/pokernet.展开更多
Text format information is full of most of the resources of Internet,which puts forward higher and higher requirements for the accuracy of text classification.Therefore,in this manuscript,firstly,we design a hybrid mo...Text format information is full of most of the resources of Internet,which puts forward higher and higher requirements for the accuracy of text classification.Therefore,in this manuscript,firstly,we design a hybrid model of bidirectional encoder representation from transformers-hierarchical attention networks-dilated convolutions networks(BERT_HAN_DCN)which based on BERT pre-trained model with superior ability of extracting characteristic.The advantages of HAN model and DCN model are taken into account which can help gain abundant semantic information,fusing context semantic features and hierarchical characteristics.Secondly,the traditional softmax algorithm increases the learning difficulty of the same kind of samples,making it more difficult to distinguish similar features.Based on this,AM-softmax is introduced to replace the traditional softmax.Finally,the fused model is validated,which shows superior performance in the accuracy rate and F1-score of this hybrid model on two datasets and the experimental analysis shows the general single models such as HAN,DCN,based on BERT pre-trained model.Besides,the improved AM-softmax network model is superior to the general softmax network model.展开更多
A discrete algorithm suitable for the computation of complex frequency-domain convolution on computers was derived. The Durbin's numerical inversion of Laplace transforms can be used to figure out the time-domain ...A discrete algorithm suitable for the computation of complex frequency-domain convolution on computers was derived. The Durbin's numerical inversion of Laplace transforms can be used to figure out the time-domain digital solution of the result of complex frequency-domain convolutions. Compared with the digital solutions and corresponding analytical solutions, it is shown that the digital solutions have high precision.展开更多
Here concerned and further investigated is a certain operator method for the computation of convolutions of polynomials.We provide a general formulation of the method with a refinement for certain old results,and also...Here concerned and further investigated is a certain operator method for the computation of convolutions of polynomials.We provide a general formulation of the method with a refinement for certain old results,and also give some new applications to convolved sums involving several noted special polynomials.The advantage of the method using operators is illustrated with concrete examples.Finally,also presented is a brief investigation on convolution polynomials having two types of summations.展开更多
Based on quantum mechanical representation and operator theory,this paper restates the two new convolutions of fractional Fourier transform(FrFT)by making full use of the conversion relationship between two mutual con...Based on quantum mechanical representation and operator theory,this paper restates the two new convolutions of fractional Fourier transform(FrFT)by making full use of the conversion relationship between two mutual conjugates:coordinate representation and momentum representation.This paper gives full play to the efficiency of Dirac notation and proves the convolutions of fractional Fourier transform from the perspective of quantum optics,a field that has been developing rapidly.These two new convolution methods have potential value in signal processing.展开更多
For a locally compact group G, L 1(G) is its group algebra and L ∞(G) is the dual of L 1(G). Lau has studied the bounded linear operators T : L ∞(G) → L ∞(G) which commute with convolutions and translations. For a...For a locally compact group G, L 1(G) is its group algebra and L ∞(G) is the dual of L 1(G). Lau has studied the bounded linear operators T : L ∞(G) → L ∞(G) which commute with convolutions and translations. For a subspace H of L ∞(G), we know that M(L ∞(G),H), the Banach algebra of all bounded linear operators on L ∞(G) into H which commute with convolutions, has been studied by Pym and Lau. In this paper, we generalize these problems to L(K)*, the dual of a hypergroup algebra L(K) in a very general setting, i. e. we do not assume that K admits a Haar measure. It should be noted that these algebras include not only the group algebra L 1(G) but also most of the semigroup algebras. Compact hypergroups have a Haar measure, however, in general it is not known that every hypergroup has a Haar measure. The lack of the Haar measure and involution presents many difficulties; however, we succeed in getting some interesting results.展开更多
Reliable traffic flow prediction is crucial for mitigating urban congestion.This paper proposes Attentionbased spatiotemporal Interactive Dynamic Graph Convolutional Network(AIDGCN),a novel architecture integrating In...Reliable traffic flow prediction is crucial for mitigating urban congestion.This paper proposes Attentionbased spatiotemporal Interactive Dynamic Graph Convolutional Network(AIDGCN),a novel architecture integrating Interactive Dynamic Graph Convolution Network(IDGCN)with Temporal Multi-Head Trend-Aware Attention.Its core innovation lies in IDGCN,which uniquely splits sequences into symmetric intervals for interactive feature sharing via dynamic graphs,and a novel attention mechanism incorporating convolutional operations to capture essential local traffic trends—addressing a critical gap in standard attention for continuous data.For 15-and 60-min forecasting on METR-LA,AIDGCN achieves MAEs of 0.75%and 0.39%,and RMSEs of 1.32%and 0.14%,respectively.In the 60-min long-term forecasting of the PEMS-BAY dataset,the AIDGCN out-performs the MRA-BGCN method by 6.28%,4.93%,and 7.17%in terms of MAE,RMSE,and MAPE,respectively.Experimental results demonstrate the superiority of our pro-posed model over state-of-the-art methods.展开更多
The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditio...The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditional approaches like network compression,quantization,and lightweight design often sacrifice accuracy or feature representation robustness.This article introduces the Fast Multi-scale Channel Shuffling Network(FMCSNet),a novel lightweight detection model optimized for mobile devices.FMCSNet integrates a fully convolutional Multilayer Perceptron(MLP)module,offering global perception without significantly increasing parameters,effectively bridging the gap between CNNs and Vision Transformers.FMCSNet achieves a delicate balance between computation and accuracy mainly by two key modules:the ShiftMLP module,including a shift operation and an MLP module,and a Partial group Convolutional(PGConv)module,reducing computation while enhancing information exchange between channels.With a computational complexity of 1.4G FLOPs and 1.3M parameters,FMCSNet outperforms CNN-based and DWConv-based ShuffleNetv2 by 1%and 4.5%mAP on the Pascal VOC 2007 dataset,respectively.Additionally,FMCSNet achieves a mAP of 30.0(0.5:0.95 IoU threshold)with only 2.5G FLOPs and 2.0M parameters.It achieves 32 FPS on low-performance i5-series CPUs,meeting real-time detection requirements.The versatility of the PGConv module’s adaptability across scenarios further highlights FMCSNet as a promising solution for real-time mobile object detection.展开更多
Sudden wildfires cause significant global ecological damage.While satellite imagery has advanced early fire detection and mitigation,image-based systems face limitations including high false alarm rates,visual obstruc...Sudden wildfires cause significant global ecological damage.While satellite imagery has advanced early fire detection and mitigation,image-based systems face limitations including high false alarm rates,visual obstructions,and substantial computational demands,especially in complex forest terrains.To address these challenges,this study proposes a novel forest fire detection model utilizing audio classification and machine learning.We developed an audio-based pipeline using real-world environmental sound recordings.Sounds were converted into Mel-spectrograms and classified via a Convolutional Neural Network(CNN),enabling the capture of distinctive fire acoustic signatures(e.g.,crackling,roaring)that are minimally impacted by visual or weather conditions.Internet of Things(IoT)sound sensors were crucial for generating complex environmental parameters to optimize feature extraction.The CNN model achieved high performance in stratified 5-fold cross-validation(92.4%±1.6 accuracy,91.2%±1.8 F1-score)and on test data(94.93%accuracy,93.04%F1-score),with 98.44%precision and 88.32%recall,demonstrating reliability across environmental conditions.These results indicate that the audio-based approach not only improves detection reliability but also markedly reduces computational overhead compared to traditional image-based methods.The findings suggest that acoustic sensing integrated with machine learning offers a powerful,low-cost,and efficient solution for real-time forest fire monitoring in complex,dynamic environments.展开更多
An image processing and deep learning method for identifying different types of rock images was proposed.Preprocessing,such as rock image acquisition,gray scaling,Gaussian blurring,and feature dimensionality reduction...An image processing and deep learning method for identifying different types of rock images was proposed.Preprocessing,such as rock image acquisition,gray scaling,Gaussian blurring,and feature dimensionality reduction,was conducted to extract useful feature information and recognize and classify rock images using Tensor Flow-based convolutional neural network(CNN)and Py Qt5.A rock image dataset was established and separated into workouts,confirmation sets,and test sets.The framework was subsequently compiled and trained.The categorization approach was evaluated using image data from the validation and test datasets,and key metrics,such as accuracy,precision,and recall,were analyzed.Finally,the classification model conducted a probabilistic analysis of the measured data to determine the equivalent lithological type for each image.The experimental results indicated that the method combining deep learning,Tensor Flow-based CNN,and Py Qt5 to recognize and classify rock images has an accuracy rate of up to 98.8%,and can be successfully utilized for rock image recognition.The system can be extended to geological exploration,mine engineering,and other rock and mineral resource development to more efficiently and accurately recognize rock samples.Moreover,it can match them with the intelligent support design system to effectively improve the reliability and economy of the support scheme.The system can serve as a reference for supporting the design of other mining and underground space projects.展开更多
With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performan...With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.展开更多
With the increasing complexity of industrial automation,planetary gearboxes play a vital role in largescale equipment transmission systems,directly impacting operational efficiency and safety.Traditional maintenance s...With the increasing complexity of industrial automation,planetary gearboxes play a vital role in largescale equipment transmission systems,directly impacting operational efficiency and safety.Traditional maintenance strategies often struggle to accurately predict the degradation process of equipment,leading to excessive maintenance costs or potential failure risks.However,existing prediction methods based on statistical models are difficult to adapt to nonlinear degradation processes.To address these challenges,this study proposes a novel condition-based maintenance framework for planetary gearboxes.A comprehensive full-lifecycle degradation experiment was conducted to collect raw vibration signals,which were then processed using a temporal convolutional network autoencoder with multi-scale perception capability to extract deep temporal degradation features,enabling the collaborative extraction of longperiod meshing frequencies and short-term impact features from the vibration signals.Kernel principal component analysis was employed to fuse and normalize these features,enhancing the characterization of degradation progression.A nonlinear Wiener process was used to model the degradation trajectory,with a threshold decay function introduced to dynamically adjust maintenance strategies,and model parameters optimized through maximum likelihood estimation.Meanwhile,the maintenance strategy was optimized to minimize costs per unit time,determining the optimal maintenance timing and preventive maintenance threshold.The comprehensive indicator of degradation trends extracted by this method reaches 0.756,which is 41.2%higher than that of traditional time-domain features;the dynamic threshold strategy reduces the maintenance cost per unit time to 55.56,which is 8.9%better than that of the static threshold optimization.Experimental results demonstrate significant reductions in maintenance costs while enhancing system reliability and safety.This study realizes the organic integration of deep learning and reliability theory in the maintenance of planetary gearboxes,provides an interpretable solution for the predictive maintenance of complex mechanical systems,and promotes the development of condition-based maintenance strategies for planetary gearboxes.展开更多
Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
In this paper,a new bolt fault diagnosis method is developed to solve the fault diagnosis problem of wind turbine flange bolts using one-dimensional depthwise separable convolutions.The main idea is to use a one-dimen...In this paper,a new bolt fault diagnosis method is developed to solve the fault diagnosis problem of wind turbine flange bolts using one-dimensional depthwise separable convolutions.The main idea is to use a one-dimensional convolutional neural network model to classify and identify the acoustic vibration signals of bolts,which represent different bolt damage states.Through the methods of knock test and modal simulation,it is concluded that the damage state of wind turbine flange bolt is related to the natural frequency distribution of acoustic vibration signal.It is found that the bolt damage state affects the modal shape of the structure,and then affects the natural frequency distribution of the bolt vibration signal.Therefore,the damage state can be obtained by identifying the natural frequency distribution of the bolt acoustic vibration signal.In the present one-dimensional depth-detachable convolutional neural network model,the one-dimensional vector is first convolved into multiple channels,and then each channel is separately learned by depth-detachable convolution,which can effectively improve the feature quality and the effect of data classification.From the perspective of the realization mechanism of convolution operation,the depthwise separable convolution operation has fewer parameters and faster computing speed,making it easier to build lightweight models and deploy them to mobile devices.展开更多
Let A be an ruth order n-dimensional tensor, where m, n are some positive integers and N ：= re（n-1）. Then A is called a Hankel tensor associated with a vector v ∈ R^N＋1 if Aσ = Vk for each k = 0,1,...,N whenever...Let A be an ruth order n-dimensional tensor, where m, n are some positive integers and N ：= re（n-1）. Then A is called a Hankel tensor associated with a vector v ∈ R^N＋1 if Aσ = Vk for each k = 0,1,...,N whenever σ= （i1,..., im） satisfies i1 ＋... ＋ im - m ＋ k. We introduce the elementary Hankel tensors which are some special Hankel tensors, and present all the eigenvalues of the elementary Hankel tensors for k = 0, 1, 2. We also show that a convolution can be expressed as the product of some third-order elementary Hankel tensors, and a Hankel tensor can be decomposed as a convolution of two Vandermonde matrices following the definition of the convolution of tensors. Finally, we use the properties of the convolution to characterize Hankel tensors and （0,1） Hankel tensors. Keywords Tensor, convolution, Hankel tensor, elementary Hankel tensor, symmetric tensor展开更多
基金supported by the European University of Atlantic.
文摘Brain tumors pose significant diagnostic challenges due to their diverse types and complex anatomical locations.Due to the increase in precision image-based diagnostic tools,driven by advancements in artificial intelligence(AI)and deep learning,there has been potential to improve diagnostic accuracy,especially with Magnetic Resonance Imaging(MRI).However,traditional state-of-the-art models lack the sensitivity essential for reliable tumor identification and segmentation.Thus,our research aims to enhance brain tumor diagnosis in MRI by proposing an advanced model.The proposed model incorporates dilated convolutions to optimize the brain tumor segmentation and classification.The proposed model is first trained and later evaluated using the BraTS 2020 dataset.In our proposed model preprocessing consists of normalization,noise reduction,and data augmentation to improve model robustness.The attention mechanism and dilated convolutions were introduced to increase the model’s focus on critical regions and capture finer spatial details without compromising image resolution.We have performed experimentation to measure efficiency.For this,we have used various metrics including accuracy,sensitivity,and curve(AUC-ROC).The proposed model achieved a high accuracy of 94%,a sensitivity of 93%,a specificity of 92%,and an AUC-ROC of 0.98,outperforming traditional diagnostic models in brain tumor detection.The proposed model accurately identifies tumor regions,while dilated convolutions enhanced the segmentation accuracy,especially for complex tumor structures.The proposed model demonstrates significant potential for clinical application,providing reliable and precise brain tumor detection in MRI.
文摘The Bernoulli convolution ν λ measure is shown to be absolutely continuous with L 2 density for almost all 12<λ<1,and singular if λ -1 is a Pisot number. It is an open question whether the Pisot type Bernoulli convolutions are the only singular ones. In this paper,we construct a family of non-Pisot type Bernoulli convolutions ν λ such that their density functions,if they exist,are not L 2. We also construct other Bernolulli convolutions whose density functions,if they exist,behave rather badly.
文摘Louis Pierre Gratiolet (1815-1865) was one of the first modern anatomists to pay attention to cerebral convolutions. Born in Sainte-Foy-la-Grande (Gironde), he moved to Paris in 1834 to study medicine, as well as comparative anatomy under Henri de Blainville (1777-1850). In 1842, he accepted de Blainville’s offer to become his assistant at the Muséum d’histoire naturelle and progressively abandoned medicine for comparative anatomy. He undertook a detailed study of brains of human and nonhuman primates and soon realized that the organizational pattern of cerebral convolutions was so predictable that it could serve as a criterion to classify primate groups. He noted that only the deepest sulci exist in lower primate forms, while the complexity of cortical folding increases markedly in great apes and humans. Gratiolet provided the first cogent description of the lobular organization of primate cerebral hemispheres. He saw the insula as a central lobe around which revolved the frontal, parietal, temporal (temporo-sphenoidal) and occipital lobes. He correctly identified most gyri and sulci on all brain surfaces, introduced the term “plis de passage” for some interconnecting gyri, and provided the first description of the optic radiations. In the early 1860s, Gratiolet fought a highly publicized battle against Paul Broca (1824-1880) on the relationship between brain and intelligence. Gratiolet agreed that the brain was most likely the seat of intelligence, but he considered human cognition far too subtle to have any direct relationship with brain size. He argued that a detailed study of the human brain architecture would be more profitable than Broca’s vain speculations on the relationship between brain weight and intelligence, which he considered a monolithic entity. Despite remarkable scientific achievements and a unique teaching capacity, Gratiolet was unable to secure any academic position until three years before his sudden death in Paris at age 49.
基金supported by National Natural Science Foundation of China(Nos.61525306,61633021,61721004,61806194,U1803261 and 61976132)Major Project for New Generation of AI(No.2018AAA0100400)+2 种基金Beijing Nova Program(No.Z201100006820079)Shandong Provincial Key Research and Development Program(No.2019JZZY010119)CAS-AIR。
文摘Pointwise convolution is usually utilized to expand or squeeze features in modern lightweight deep models.However,it takes up most of the overall computational cost(usually more than 90%).This paper proposes a novel Poker module to expand features by taking advantage of cheap depthwise convolution.As a result,the Poker module can greatly reduce the computational cost,and meanwhile generate a large number of effective features to guarantee the performance.The proposed module is standardized and can be employed wherever the feature expansion is needed.By varying the stride and the number of channels,different kinds of bottlenecks are designed to plug the proposed Poker module into the network.Thus,a lightweight model can be easily assembled.Experiments conducted on benchmarks reveal the effectiveness of our proposed Poker module.And our Poker Net models can reduce the computational cost by 7.1%-15.6%.Poker Net models achieve comparable or even higher recognition accuracy than previous state-of-the-art(SOTA)models on the Image Net ILSVRC2012 classification dataset.Code is available at https://github.com/diaomin/pokernet.
基金Fundamental Research Funds for the Central University,China(No.2232018D3-17)。
文摘Text format information is full of most of the resources of Internet,which puts forward higher and higher requirements for the accuracy of text classification.Therefore,in this manuscript,firstly,we design a hybrid model of bidirectional encoder representation from transformers-hierarchical attention networks-dilated convolutions networks(BERT_HAN_DCN)which based on BERT pre-trained model with superior ability of extracting characteristic.The advantages of HAN model and DCN model are taken into account which can help gain abundant semantic information,fusing context semantic features and hierarchical characteristics.Secondly,the traditional softmax algorithm increases the learning difficulty of the same kind of samples,making it more difficult to distinguish similar features.Based on this,AM-softmax is introduced to replace the traditional softmax.Finally,the fused model is validated,which shows superior performance in the accuracy rate and F1-score of this hybrid model on two datasets and the experimental analysis shows the general single models such as HAN,DCN,based on BERT pre-trained model.Besides,the improved AM-softmax network model is superior to the general softmax network model.
文摘A discrete algorithm suitable for the computation of complex frequency-domain convolution on computers was derived. The Durbin's numerical inversion of Laplace transforms can be used to figure out the time-domain digital solution of the result of complex frequency-domain convolutions. Compared with the digital solutions and corresponding analytical solutions, it is shown that the digital solutions have high precision.
文摘Here concerned and further investigated is a certain operator method for the computation of convolutions of polynomials.We provide a general formulation of the method with a refinement for certain old results,and also give some new applications to convolved sums involving several noted special polynomials.The advantage of the method using operators is illustrated with concrete examples.Finally,also presented is a brief investigation on convolution polynomials having two types of summations.
基金National Natural Science Foundation of China(Grant Number:11304126)College Students' Innovation Training Program(Grant Number:202110299696X)。
文摘Based on quantum mechanical representation and operator theory,this paper restates the two new convolutions of fractional Fourier transform(FrFT)by making full use of the conversion relationship between two mutual conjugates:coordinate representation and momentum representation.This paper gives full play to the efficiency of Dirac notation and proves the convolutions of fractional Fourier transform from the perspective of quantum optics,a field that has been developing rapidly.These two new convolution methods have potential value in signal processing.
文摘For a locally compact group G, L 1(G) is its group algebra and L ∞(G) is the dual of L 1(G). Lau has studied the bounded linear operators T : L ∞(G) → L ∞(G) which commute with convolutions and translations. For a subspace H of L ∞(G), we know that M(L ∞(G),H), the Banach algebra of all bounded linear operators on L ∞(G) into H which commute with convolutions, has been studied by Pym and Lau. In this paper, we generalize these problems to L(K)*, the dual of a hypergroup algebra L(K) in a very general setting, i. e. we do not assume that K admits a Haar measure. It should be noted that these algebras include not only the group algebra L 1(G) but also most of the semigroup algebras. Compact hypergroups have a Haar measure, however, in general it is not known that every hypergroup has a Haar measure. The lack of the Haar measure and involution presents many difficulties; however, we succeed in getting some interesting results.
文摘Reliable traffic flow prediction is crucial for mitigating urban congestion.This paper proposes Attentionbased spatiotemporal Interactive Dynamic Graph Convolutional Network(AIDGCN),a novel architecture integrating Interactive Dynamic Graph Convolution Network(IDGCN)with Temporal Multi-Head Trend-Aware Attention.Its core innovation lies in IDGCN,which uniquely splits sequences into symmetric intervals for interactive feature sharing via dynamic graphs,and a novel attention mechanism incorporating convolutional operations to capture essential local traffic trends—addressing a critical gap in standard attention for continuous data.For 15-and 60-min forecasting on METR-LA,AIDGCN achieves MAEs of 0.75%and 0.39%,and RMSEs of 1.32%and 0.14%,respectively.In the 60-min long-term forecasting of the PEMS-BAY dataset,the AIDGCN out-performs the MRA-BGCN method by 6.28%,4.93%,and 7.17%in terms of MAE,RMSE,and MAPE,respectively.Experimental results demonstrate the superiority of our pro-posed model over state-of-the-art methods.
基金funded by the National Natural Science Foundation of China under Grant No.62371187the Open Program of Hunan Intelligent Rehabilitation Robot and Auxiliary Equipment Engineering Technology Research Center under Grant No.2024JS101.
文摘The ubiquity of mobile devices has driven advancements in mobile object detection.However,challenges in multi-scale object detection in open,complex environments persist due to limited computational resources.Traditional approaches like network compression,quantization,and lightweight design often sacrifice accuracy or feature representation robustness.This article introduces the Fast Multi-scale Channel Shuffling Network(FMCSNet),a novel lightweight detection model optimized for mobile devices.FMCSNet integrates a fully convolutional Multilayer Perceptron(MLP)module,offering global perception without significantly increasing parameters,effectively bridging the gap between CNNs and Vision Transformers.FMCSNet achieves a delicate balance between computation and accuracy mainly by two key modules:the ShiftMLP module,including a shift operation and an MLP module,and a Partial group Convolutional(PGConv)module,reducing computation while enhancing information exchange between channels.With a computational complexity of 1.4G FLOPs and 1.3M parameters,FMCSNet outperforms CNN-based and DWConv-based ShuffleNetv2 by 1%and 4.5%mAP on the Pascal VOC 2007 dataset,respectively.Additionally,FMCSNet achieves a mAP of 30.0(0.5:0.95 IoU threshold)with only 2.5G FLOPs and 2.0M parameters.It achieves 32 FPS on low-performance i5-series CPUs,meeting real-time detection requirements.The versatility of the PGConv module’s adaptability across scenarios further highlights FMCSNet as a promising solution for real-time mobile object detection.
基金funded by the Directorate of Research and Community Service,Directorate General of Research and Development,Ministry of Higher Education,Science and Technologyin accordance with the Implementation Contract for the Operational Assistance Program for State Universities,Research Program Number:109/C3/DT.05.00/PL/2025.
文摘Sudden wildfires cause significant global ecological damage.While satellite imagery has advanced early fire detection and mitigation,image-based systems face limitations including high false alarm rates,visual obstructions,and substantial computational demands,especially in complex forest terrains.To address these challenges,this study proposes a novel forest fire detection model utilizing audio classification and machine learning.We developed an audio-based pipeline using real-world environmental sound recordings.Sounds were converted into Mel-spectrograms and classified via a Convolutional Neural Network(CNN),enabling the capture of distinctive fire acoustic signatures(e.g.,crackling,roaring)that are minimally impacted by visual or weather conditions.Internet of Things(IoT)sound sensors were crucial for generating complex environmental parameters to optimize feature extraction.The CNN model achieved high performance in stratified 5-fold cross-validation(92.4%±1.6 accuracy,91.2%±1.8 F1-score)and on test data(94.93%accuracy,93.04%F1-score),with 98.44%precision and 88.32%recall,demonstrating reliability across environmental conditions.These results indicate that the audio-based approach not only improves detection reliability but also markedly reduces computational overhead compared to traditional image-based methods.The findings suggest that acoustic sensing integrated with machine learning offers a powerful,low-cost,and efficient solution for real-time forest fire monitoring in complex,dynamic environments.
基金financially supported by the National Science and Technology Major Project——Deep Earth Probe and Mineral Resources Exploration(No.2024ZD1003701)the National Key R&D Program of China(No.2022YFC2905004)。
文摘An image processing and deep learning method for identifying different types of rock images was proposed.Preprocessing,such as rock image acquisition,gray scaling,Gaussian blurring,and feature dimensionality reduction,was conducted to extract useful feature information and recognize and classify rock images using Tensor Flow-based convolutional neural network(CNN)and Py Qt5.A rock image dataset was established and separated into workouts,confirmation sets,and test sets.The framework was subsequently compiled and trained.The categorization approach was evaluated using image data from the validation and test datasets,and key metrics,such as accuracy,precision,and recall,were analyzed.Finally,the classification model conducted a probabilistic analysis of the measured data to determine the equivalent lithological type for each image.The experimental results indicated that the method combining deep learning,Tensor Flow-based CNN,and Py Qt5 to recognize and classify rock images has an accuracy rate of up to 98.8%,and can be successfully utilized for rock image recognition.The system can be extended to geological exploration,mine engineering,and other rock and mineral resource development to more efficiently and accurately recognize rock samples.Moreover,it can match them with the intelligent support design system to effectively improve the reliability and economy of the support scheme.The system can serve as a reference for supporting the design of other mining and underground space projects.
基金supported by the Natural Science Foundation of China No.62362008the Major Scientific and Technological Special Project of Guizhou Province([2024]014).
文摘With the rapid development of the Artificial Intelligence of Things(AIoT),convolutional neural networks(CNNs)have demonstrated potential and remarkable performance in AIoT applications due to their excellent performance in various inference tasks.However,the users have concerns about privacy leakage for the use of AI and the performance and efficiency of computing on resource-constrained IoT edge devices.Therefore,this paper proposes an efficient privacy-preserving CNN framework(i.e.,EPPA)based on the Fully Homomorphic Encryption(FHE)scheme for AIoT application scenarios.In the plaintext domain,we verify schemes with different activation structures to determine the actual activation functions applicable to the corresponding ciphertext domain.Within the encryption domain,we integrate batch normalization(BN)into the convolutional layers to simplify the computation process.For nonlinear activation functions,we use composite polynomials for approximate calculation.Regarding the noise accumulation caused by homomorphic multiplication operations,we realize the refreshment of ciphertext noise through minimal“decryption-encryption”interactions,instead of adopting bootstrapping operations.Additionally,in practical implementation,we convert three-dimensional convolution into two-dimensional convolution to reduce the amount of computation in the encryption domain.Finally,we conduct extensive experiments on four IoT datasets,different CNN architectures,and two platforms with different resource configurations to evaluate the performance of EPPA in detail.
基金funded by scientific research projects under Grant JY2024B011.
文摘With the increasing complexity of industrial automation,planetary gearboxes play a vital role in largescale equipment transmission systems,directly impacting operational efficiency and safety.Traditional maintenance strategies often struggle to accurately predict the degradation process of equipment,leading to excessive maintenance costs or potential failure risks.However,existing prediction methods based on statistical models are difficult to adapt to nonlinear degradation processes.To address these challenges,this study proposes a novel condition-based maintenance framework for planetary gearboxes.A comprehensive full-lifecycle degradation experiment was conducted to collect raw vibration signals,which were then processed using a temporal convolutional network autoencoder with multi-scale perception capability to extract deep temporal degradation features,enabling the collaborative extraction of longperiod meshing frequencies and short-term impact features from the vibration signals.Kernel principal component analysis was employed to fuse and normalize these features,enhancing the characterization of degradation progression.A nonlinear Wiener process was used to model the degradation trajectory,with a threshold decay function introduced to dynamically adjust maintenance strategies,and model parameters optimized through maximum likelihood estimation.Meanwhile,the maintenance strategy was optimized to minimize costs per unit time,determining the optimal maintenance timing and preventive maintenance threshold.The comprehensive indicator of degradation trends extracted by this method reaches 0.756,which is 41.2%higher than that of traditional time-domain features;the dynamic threshold strategy reduces the maintenance cost per unit time to 55.56,which is 8.9%better than that of the static threshold optimization.Experimental results demonstrate significant reductions in maintenance costs while enhancing system reliability and safety.This study realizes the organic integration of deep learning and reliability theory in the maintenance of planetary gearboxes,provides an interpretable solution for the predictive maintenance of complex mechanical systems,and promotes the development of condition-based maintenance strategies for planetary gearboxes.
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
基金supported in part by the National Key R&D Program of China(Nos.2021YFE0206100 and 2018YFB1702300)the National Natural Science Foundation of China(No.62073321)+1 种基金the National Defense Basic Scientific Research Program(No.JCKY2019203C029)the Science and Technology Development Fund,Macao SAR(No.0015/2020/AMJ).
文摘In this paper,a new bolt fault diagnosis method is developed to solve the fault diagnosis problem of wind turbine flange bolts using one-dimensional depthwise separable convolutions.The main idea is to use a one-dimensional convolutional neural network model to classify and identify the acoustic vibration signals of bolts,which represent different bolt damage states.Through the methods of knock test and modal simulation,it is concluded that the damage state of wind turbine flange bolt is related to the natural frequency distribution of acoustic vibration signal.It is found that the bolt damage state affects the modal shape of the structure,and then affects the natural frequency distribution of the bolt vibration signal.Therefore,the damage state can be obtained by identifying the natural frequency distribution of the bolt acoustic vibration signal.In the present one-dimensional depth-detachable convolutional neural network model,the one-dimensional vector is first convolved into multiple channels,and then each channel is separately learned by depth-detachable convolution,which can effectively improve the feature quality and the effect of data classification.From the perspective of the realization mechanism of convolution operation,the depthwise separable convolution operation has fewer parameters and faster computing speed,making it easier to build lightweight models and deploy them to mobile devices.
文摘Let A be an ruth order n-dimensional tensor, where m, n are some positive integers and N ：= re（n-1）. Then A is called a Hankel tensor associated with a vector v ∈ R^N＋1 if Aσ = Vk for each k = 0,1,...,N whenever σ= （i1,..., im） satisfies i1 ＋... ＋ im - m ＋ k. We introduce the elementary Hankel tensors which are some special Hankel tensors, and present all the eigenvalues of the elementary Hankel tensors for k = 0, 1, 2. We also show that a convolution can be expressed as the product of some third-order elementary Hankel tensors, and a Hankel tensor can be decomposed as a convolution of two Vandermonde matrices following the definition of the convolution of tensors. Finally, we use the properties of the convolution to characterize Hankel tensors and （0,1） Hankel tensors. Keywords Tensor, convolution, Hankel tensor, elementary Hankel tensor, symmetric tensor
