检索结果-南通市图书馆

NASTCNN: a spatio-temporal convolutional neural network with non-local attention mechanism for image-induced brain activity pattern recognition

引用

Optoelectronics Letters 2025年

作者： TAN Xiangxin SHAO Mingxiong ZENG Ke GU Yue School of Computer Science and Engineering， Tianjin University of technology Engineering Research Center of learning-Based Intelligent System (Ministry of Education)， Tianjin University of technology Krembil Research Institute， University Health Network

Deciphering the brain’s visual processing mechanisms has always been a hot topic. Using neural signal patterns to predict the corresponding external visual stimuli is crucial in understanding the interaction of real visual scenes. There are some challenges in the study of image-induced brain activity pattern recognition based on electroencephalogram (EEG)， e.g. how to fully utilize the effective features of EEG signals to improve the classification， and how the frequency and the length of EEG effect the model performance. We design an experimental paradigm that minimizes the interaction between visual stimuli， and propose a spatio-temporal convolutional neural network with non-local attention mechanism (NASTCNN) for image-induced brain activity pattern recognition. The framework utilizes non-local feature extraction operations to realize global information interaction between EEG signals over long distances and enhance the feature expression of some important information. Our model achieves the best performance on both MNIST-EEG and ImageNet-EEG datasets. The classification results are 24.00% and 50.10% respectively， higher than the random probability. We also analyze the effects of different EEG frequency bands and time subsequences of different lengths on the classification performance and which brain regions are activated over time. The results show that when full frequency band and entire length sequence are selected， the classification result is optimal. The occipital region is essential for processing visual tasks， and the temporal region and some frontal regions also play important role in the later stage of visual decoding. This research can help us better understand the visual processing mechanism of the brain.

关键词：

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

MoS2 synaptic transistor with one-step manufacture

引用

Science China(Information Sciences) 2025年第1期68卷 357-365页

作者： Yihao GUO Yang WANG Wenjie DENG Yi WU Jingtao LI Kexin LI Yuehui ZHAO Songlin YU Xiaoting WANG Yongzhe ZHANG Hui YAN Key Laboratory of Optoelectronics technology of Education Ministry of China Faculty of Information TechnologyBeijing University of Technology Key Laboratory of Advanced Functional Materials Ministry of Education Faculty of Materials and ManufacturingBeijing University of Technology The 11th Research Institute of Electronics technology Group Corporation

The synaptic transistor,which is a neuromorphic device with brain-like functions,has garnered significant attention due to characteristics such as low energy consumption,high integration,multiple functions,and high computational ***,the current synaptic transistor possesses a complicated structure and involves complex manufacturing steps,which is not conducive to low-cost,large-scale bionic ***,we introduce a simple MoS2synaptic transistor exhibiting robust synaptic *** transistor utilizes magnetron sputtering manufacture to combine the process of device metal electrode deposition and the introduction of defect states,which helps streamline the manufacturing process into a single *** long-term potentiation and long-term depression in photoelectric coordination are initially ***,biological functions,such as double pulse facilitation and the transformation behavior of short-term memory to long-term memory,are thoroughly evaluated,along with learning-forgetting-relearning ***,the device exhibits robust synaptic plasticity until incident light angles down to 30° to imitate the multi-to-one axon-cellular ***,a classical simulation of Pavlov's dog conditioned reflex is demonstrated along with the multiview extraction *** experimental results demonstrate a promising synaptic transistor with a simple structure,manufacturing technique,and excellent *** further indicates the potential for large-scale integration of bionic neuromorphic systems.

关键词： process optimization artificial synaptic device one-stepmanufacture synaptic plasticity bionic simulation

来源：

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

learning to see speckle in the weak laser field through multimode fiber

引用

Optoelectronics Letters 2025年

作者： JI Yunqi SONG Binbin LI Xueqing LI Yonghui The Engineering Research Center of learning-Based Intelligent System (Ministry of Education)， the Key Laboratory of Computer Vision and Systems (Ministry of Education)， and the School of Computer Science and Engineering， Tianjin University of technology

Multimode fibers (MMFs) have great potential for endoscopic imaging due to the high number of modes and a small core diameter. Deep learning based on neural networks has received increasing attention in the field of scattering image reconstruction. However， most studies focus on designing complex network architectures to improve reconstruction， but these network models struggle to reconstruct images in a weak laser field. In the paper， a lightweight generative adversarial network model combined with a histogram specification algorithm is designed to reconstruct speckles in the weak laser field through MMF. Experimental results show that the reconstruction results of our algorithm have better metrics. Moreover， the model demonstrates excellent cross-domain generalization ability with regards to the Fashion-MNIST dataset. It is worth mentioning that we found that the speckles after inactivation still retain the ability to be reconstructed， which enhances the robustness of the model

关键词：

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Few-shot learning for screening 2D Ga2CoS4-x supported single-atom catalysts for hydrogen production

引用

Journal of Energy Chemistry 2025年第1期100卷 665-673页

作者： Nabil Khossossi Poulumi Dey Department of Mechanical Engineering (ME) Delft University of Technology

Hydrogen generation and related energy applications heavily rely on the hydrogen evolution reaction（HER）,which faces challenges of slow kinetics and high *** electrocatalysts,particularly single-atom catalysts （SACs） on two-dimensional （2D） materials,are *** study presents a few-shot machine learning （ML） assisted high-throughput screening of 2D septuple-atomic-layer Ga2CoS4-xsupported SACs to predict HER catalytic ***,density functional theory （DFT）calculations showed that 2D Ga2CoS4is inactive for ***,defective Ga2CoS4-x（x=0–0.25）monolayers exhibit excellent HER activity due to surface sulfur vacancies （SVs）,with predicted overpotentials （0–60 mV） comparable to or lower than commercial Pt/C,which typically exhibits an overpotential of around 50 m V in the acidic electrolyte,when the concentration of surface SV is lower than 8.3%.SVs generate spin-polarized states near the Fermi level,making them effective HER *** demonstrate ML-accelerated HER overpotential predictions for all transition metal SACs on 2D *** DFT data from 18 SACs,an ML model with high prediction accuracy and reduced computation time was *** intrinsic descriptor linking SAC atomic properties to HER overpotential was *** study thus provides a framework for screening SACs on 2D materials,enhancing catalyst design.

关键词： Hydrogen production Electrocatalyst 2D material Density functional theory Machine learning Surface sulfur vacancy

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Grid-Connected/Islanded Switching Control Strategy for Photovoltaic Storage Hybrid Inverters Based on Modified Chimpanzee Optimization Algorithm

引用

Energy Engineering 2025年第1期122卷 265-284页

作者： Chao Zhou NarisuWang Fuyin Ni Wenchao Zhang School of Electrical and Information Engineering Jiangsu University of TechnologyChangzhou213001China Jiangsu Key Laboratory of Power Transmission&Distribution Equipment technology Jiangsu University of TechnologyChangzhou213001China

Uneven power distribution,transient voltage,and frequency deviations are observed in the photovoltaic storage hybrid inverter during the switching between grid-connected and island *** response to these issues,this paper proposes a grid-connected/island switching control strategy for photovoltaic storage hybrid inverters based on the modified chimpanzee optimization *** proposed strategy incorporates coupling compensation and power differentiation elements based on the traditional droop ***,it combines the angular frequency and voltage amplitude adjustments provided by the phase-locked loop-free pre-synchronization control *** pre-synchronization is achieved by regulating the virtual current to zero and aligning the photovoltaic storage hybrid inverter with the grid ***,two novel operators,learning and emotional behaviors are introduced to enhance the optimization precision of the chimpanzee *** operators ensure high-precision and high-reliability optimization of the droop control parameters for photovoltaic storage hybrid inverters.A Simulink model was constructed for simulation analysis,which validated the optimized control strategy’s ability to evenly distribute power under load *** strategy effectively mitigated transient voltage and current surges during mode ***,seamless and efficient switching between gridconnected and island modes was achieved for the photovoltaic storage hybrid *** enhanced energy utilization efficiency,in turn,offers robust technical support for grid stability.

关键词： Photovoltaic storage hybrid inverters modified chimpanzee optimization algorithm droop control seamless switching

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Electrochemical sensor for dopamine detection based on multiwalled carbon nanotube/molybdenum disulfide quantum dots with machine learning integration and anti-interference capability

引用

Tungsten 2025年第1期7卷 137-151页

作者： Geethukrishnan Onkar Apte Kiran Kumar Tadi Energy Electrodics and Electrocatalysis (EEE) Lab Department of Chemistry School of Advanced Sciences Vellore Institute of Technology School of Computer Science and Engineering Vellore Institute of Technology

The precise detection and measurement of dopamine（DA）,a crucial neurotransmitter in the human body,plays a significant role in diagnosing,preventing,and treating neurological diseases associated with its levels.A highly sensitive DA electrochemical sensor was constructed by combining molybdenum disulfide quantum dots（MSQDs） with multiwalled carbon nanotubes（MWCNTs）.The MSQDs were synthesized using the shear exfoliation *** sensors consist of MSQDs with Mo-S edge catalytic centers for the DA redox reaction,and MWCNTs amplify the sensor *** linearity of the sensor for the detection of DA was tested in the presence of ascorbic acid(AA,50 μmol·L-1) and uric acid(UA,200 μmol·L-1),and exhibited linearity from 2 to 966 μmol·L-1of DA with 0.097 μA(mol·L-1)-1sensitivity and a low limit of detection of0.6 μmol·L-1（the ratio between signal and noise,S/N=3）.Moreover,the sensitivity and selectivity of the sensor were also studied using *** is no increase in amperometric current after adding the most potentially interfering *** sensor was successfully applied to recover DA in human blood sera ***,machine learning algorithms were operated to aid in the near-precise detection of DA in the heterogeneous mixture containing AA and *** algorithms facilitate the identification and quantification of DA amidst coexisting interferents,including AA,that are commonly present in biological matrices.

关键词： Biosensor Complex mixture Dopamine Machine learning MoS2 Multiwalled carbon nanotubes

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Domain generalization with semi-supervised learning for people-centric activity recognition

引用

Science China(Information Sciences) 2025年第1期68卷 171-188页

作者： Jing LIU Wei ZHU Di LI Xing HU Liang SONG Academy for Engineering & technology Fudan University Shanghai East-bund Research Institute on Networking Systems of AI School of Optoelectronic Information and Computer Engineering University of Shanghai for Science & Technology

People-centric activity recognition is one of the most critical technologies in a wide range of real-world applications,including intelligent transportation systems, healthcare services, and brain-computer interfaces. Large-scale data collection and annotation make the application of machine learning algorithms prohibitively expensive when adapting to new tasks. One way of circumventing this limitation is to train the model in a semi-supervised learning manner that utilizes a percentage of unlabeled data to reduce the labeling burden in prediction tasks. Despite their appeal, these models often assume that labeled and unlabeled data come from similar distributions, which leads to the domain shift problem caused by the presence of distribution gaps. To address these limitations, we propose herein a novel method for people-centric activity recognition,called domain generalization with semi-supervised learning(DGSSL), that effectively enhances the representation learning and domain alignment capabilities of a model. We first design a new autoregressive discriminator for adversarial training between unlabeled and labeled source domains, extracting domain-specific features to reduce the distribution gaps. Second, we introduce two reconstruction tasks to capture the task-specific features to avoid losing information related to representation learning while maintaining task-specific consistency. Finally, benefiting from the collaborative optimization of these two tasks, the model can accurately predict both the domain and category labels of the source domains for the classification task. We conduct extensive experiments on three real-world sensing datasets. The experimental results show that DGSSL surpasses the three state-of-the-art methods with better performance and generalization.

关键词： activity recognition deep learning domain generalization semi-supervised learning adversarial training

来源：

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Fatigue Strength Analysis of Dissimilar Aluminum Alloy TIG Welds

引用

Journal of Wuhan University of technology(Materials Science) 2025年第1期40卷 265-274页

作者： LIAO Xiangyun WANG Ruijie LIU Guoshou ZHAO Pinglin Faculty of Mechanical and Electrical Engineering Kunming University of Science and TechnologyKunming 650500China Faculty of Civil Engineering and Mechanics Kunming University of Science and TechnologyKunming 650500China

The constant amplitude loading fatigue tests were carried out on the 6061/7075 aluminum alloy TIG fillet welded lap specimens in this study,and the weld seam cross-section hardness was *** experimental results show that most specimens mainly failed at the 7075 side weld toes even though the base material tensile strength of 7075 is higher than that of *** maximum stress-strain concentration in the two finite element models is located at the 7075 side weld toe,which is basically consistent with the actual fracture *** weld zone on the 7075 side experiences severe material softening,with a large ***,the Vickers hardness value on the 6061 side negligibly changes and fluctuates around 70 *** obvious defects are found on the fatigue fracture,but a large number of secondary cracks *** germinate from the weld toe and propagate in the direction of the plate *** reinforcement has a serious impact on fatigue *** life will decrease exponentially as the weld reinforcement increases under low *** is found that the notch stress method can give a better fatigue life prediction for TIG weldments,and the errors of the predicted results are within the range of two factors,while the prediction accuracy decreases under low *** equivalent structural stress method can also be used for fatigue life prediction of TIG weldments,but the errors of prediction results are within the range of three factors,and the accuracy decreases under high stress.

关键词： TIG welding notch stress method equivalent structural stress method fatigue life finite element analysis

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Targets capture by distributed active swarms via bio-inspired reinforcement learning

引用

Science China(Physics,Mechanics & Astronomy) 2025年第1期Mechanics & Astronomy) .卷 210-221页

作者： Kun Xu Yue Li Jun Sun Shuyuan Du Xinpeng Di Yuguang Yang Bo Li Institute of Biomechanics and Medical Engineering Applied Mechanics Laboratory Department of Engineering MechanicsTsinghua University Institute of Nuclear and New Energy technology Tsinghua University Shanghai Aerospace Control technology Institute Shanghai Key Laboratory of Aerospace Intelligent Control technology College of Automation Engineering Nanjing University of Aeronautics and Astronautics

Natural swarms arranged from cells to herds are usually decentralized but display intriguing collective intelligence in coordinating individuals across large scales to efficiently achieve their common goals. learning from nature may provide new strategies for controlling collective dynamics of synthetic swarms to accomplish specific functions. Here, we present a bioinspired computational framework that steers distributed active swarms to collectively capture and merge targets via reinforcement learning. We exploit collective milling structures of natural herds to cage the targets, and adopt a switching control policy inspired by sperms' chiral dynamics to optimize the trajectories of individuals, through which the active swarms can selforganize to enclose single or multiple distant targets in a dynamical, adaptive and scalable manner. There exists a critical swarm size, beyond which the excessive competition between agents would generate large mechanical forces, leading to capture instability but enabling the transition from short-distance to long-distance merging capture of multiple targets. This work provides physical insights into distributed active swarms and could offer a multilevel, decentralized strategy toward controlling swarm robotics in wide applications such as bio-medical devices, machine immunity, and target clearance.

关键词： swarm intelligence robotics multi-tasks artificial intelligence bionics

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Stochastic Augmented-Based Dual-Teaching for Semi-Supervised Medical Image Segmentation

引用

Computers, Materials & Continua 2025年第1期82卷 543-560页

作者： Hengyang Liu Yang Yuan Pengcheng Ren Chengyun Song Fen Luo School of Computer Science and Engineering Chongqing University of TechnologyChongqing400054China College of Artificial Intelligence Chongqing Technology and Business UniversityChongqing400067China

Existing semi-supervisedmedical image segmentation algorithms use copy-paste data augmentation to correct the labeled-unlabeled data distribution ***,current copy-paste methods have three limitations:(1)training the model solely with copy-paste mixed pictures from labeled and unlabeled input loses a lot of labeled information;(2)low-quality pseudo-labels can cause confirmation bias in pseudo-supervised learning on unlabeled data;(3)the segmentation performance in low-contrast and local regions is less than *** design a Stochastic Augmentation-Based Dual-Teaching Auxiliary Training Strategy(SADT),which enhances feature diversity and learns high-quality features to overcome these *** be more precise,SADT trains the Student Network by using pseudo-label-based training from Teacher Network 1 and supervised learning with labeled data,which prevents the loss of rare labeled *** introduce a bi-directional copy-pastemask with progressive high-entropy filtering to reduce data distribution disparities and mitigate confirmation bias in *** the mixed images,Deep-Shallow Spatial Contrastive learning(DSSCL)is proposed in the feature spaces of Teacher Network 2 and the Student Network to improve the segmentation capabilities in low-contrast and local *** this procedure,the features retrieved by the Student Network are subjected to a random feature perturbation *** two openly available datasets,extensive trials show that our proposed SADT performs much better than the state-ofthe-art semi-supervised medical segmentation *** only 10%of the labeled data for training,SADT was able to acquire a Dice score of 90.10%on the ACDC(Automatic Cardiac Diagnosis Challenge)dataset.

关键词： Semi-supervised medical image segmentation contrastive learning stochastic augmented

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：