检索结果-南通市图书馆

Chinese Journal of Electronics 2021年第2期30卷 289-295页

作者： LI Youjiao ZHUO Li LI Jiafeng ZHANG Jing beijing key laboratory of computational intelligence and intelligent System Beijing University of Technology Faculty of Information Technology College of Micro-electronics Beijing University of Technology College of Computer Science and Technology Shandong University of Technology

A two-level hierarchical scheme for video-based person re-identification(re-id) is presented, with the aim of learning a pedestrian appearance model through more complete walking cycle extraction. Specifically,given a video with consecutive frames, the objective of the first level is to detect the key frame with lightweight Convolutional neural network(CNN) of PCANet to reflect the summary of the video content. At the second level, on the basis of the detected key frame, the pedestrian walking cycle is extracted from the long video sequence. Moreover,local features of Local maximal occurrence(LOMO) of the walking cycle are extracted to represent the pedestrian’s appearance information. In contrast to the existing walking-cycle-based person re-id approaches, the proposed scheme relaxes the limit on step number for a walking cycle, thus making it flexible and less affected by noisy frames. Experiments are conducted on two benchmark datasets: PRID 2011 and i LIDS-VID. The experimental results demonstrate that our proposed scheme outperforms the six state-of-art video-based re-id methods, and is more robust to the severe video noises and variations in pose,lighting, and camera viewpoint.

关键词： Image recognition Computer vision and image processing techniques Video signal processing Neural nets Video-based person re-identification Convolutional neural network key frame detection Walking cycle extraction

Attention-Based Multi-Scale Prediction Network for Time-Series Data

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

China Communications 2022年第5期19卷 286-301页

作者： Junjie Li Lin Zhu Yong Zhang Da Guo Xingwen Xia School of Electronic Engineering Beijing University of Posts and TelecommunicationsBeijing 100876China beijing key laboratory of Work Safety intelligent Monitoring(beijing University of Posts and Telecommunications) Beijing 100876China China Mobile Research Institute Beijing 100053China

Time series data is a kind of data accumulated over time,which can describe the change of *** kind of data reflects the degree of change of a certain thing or *** existing technologies such as LSTM and ARIMA are better than convolutional neural network in time series prediction,but they are not enough to mine the periodicity of *** this article,we perform periodic analysis on two types of time series data,select time metrics with high periodic characteristics,and propose a multi-scale prediction model based on the attention mechanism for the periodic trend of the data.A loss calculation method for traffic time series characteristics is proposed as *** experiments have been conducted on actual data *** experiments show that the method proposed in this paper has better performance than commonly used traffic prediction methods(ARIMA,LSTM,etc.)and 3%-5%increase on MAPE.

关键词： network traffic prediction attention mechanism neural network machine learning single point forecast

A Spatial Cognitive Model that Integrates the Effects of Endogenous and Exogenous Information on the Hippocampus and Striatum

在线全文

学校读者我要写书评

暂无评论

International Journal of Automation and computing 2021年第4期18卷 632-644页

作者： Jing Huang He-Yuan Yang Xiao-Gang Ruan Nai-Gong Yu Guo-Yu Zuo Hao-Meng Liu Faculty of Information Technology Beijing University of TechnologyBeijing 100124China beijing key laboratory of computational intelligence and intelligence System Beijing 100124China

Reproducing the spatial cognition of animals using computational models that make agents navigate autonomously has attracted much attention. Many biologically inspired models for spatial cognition focus mainly on the simulation of the hippocampus and only consider the effect of external environmental information(i.e., exogenous information) on the hippocampal coding. However, neurophysiological studies have shown that the striatum, which is closely related to the hippocampus, also plays an important role in spatial cognition and that information inside animals(i.e., endogenous information) also affects the encoding of the hippocampus. Inspired by the progress made in neurophysiological studies, we propose a new spatial cognitive model that consists of analogies between the hippocampus and striatum. This model takes into consideration how both exogenous and endogenous information affects coding by the environment. We carried out a series of navigation experiments that simulated a water maze and compared our model with other models. Our model is self-adaptable and robust and has better performance in navigation path length. We also discuss the possible reasons for the results and how our findings may help us understand real mechanisms in the spatial cognition of animals.

关键词： Exogenous and endogenous information hippocampus striatum spatial cognition brain-inspired computation

Temporal-spatial dynamic characteristics of vehicle emissions on intercity roads in Guangdong Province based on vehicle identity detection data

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2023年第8期130卷 126-138页

作者： Hui Ding Yongming Zhao Shenhua Miao Tong Chen Yonghong Liu College of Automation&College of Artifical intelligence Nanjing University of Posts and TelecommunicationsNanjing 210023China School of intelligent systems Engineering Sun Yat-Sen UniversityGuangzhou 510275China Guangdong Provincial key laboratory of intelligent Transport System Guangzhou 510275China Guangdong Provincial Engineering Research Center for Traffic Environmental Monitoring and Control Guangzhou 510275China

Estimating intercity vehicle emissions precisely would benefit collaborative control in multiple *** the variability of emissions caused by vehicles,roads,and traffic,the 24-hour change characteristics of air pollutants(CO,HC,NO_(X),PM_(2.5))on the intercity road network of Guangdong Province by vehicle categories and road links were revealed based on vehicle identity detection data in real-life traffic for each hour in July *** results showed that the spatial diversity of emissions caused by the unbalanced economywas *** vehicle emissions in the Pearl River Delta region(PRD)with a higher economic level were approximately 1–2 times those in the non-Pearl RiverDelta region(non-PRD).Provincial roads with high loads became potential sources of high ***,emission control policies must emphasize the PRD and key roads by travel guidance to achieve greater *** passenger cars with a large proportion of traffic dominated morning and evening peaks in the 24-hour period and were the dominant contributors to CO and HC emissions,contributing more than 50%in the daytime(7:00–23:00)and higher than 26%at night(0:00–6:00).Diesel trucks made up 10%of traffic,but were the dominant player at night,contributed 50%–90%to NO_(X) and PM_(2.5) emissions,with amarked 24-hour change rule of more than 80%at night(23:00–5:00)and less than 60%during ***,targeted control measures by time-section should be set up on collaborative *** findings provide time-varying decision support for variable vehicle emission control on a large scale.

关键词： Intercity roads Dynamic vehicle emissions Vehicle identity detection data Diesel trucks

Social relation and physical lane aggregator:integrating social and physical features for multimodal motion prediction

在线全文

学校读者我要写书评

暂无评论

Journal of intelligent and Connected Vehicles 2022年第3期5卷 302-308页

作者： Qiyuan Chen Zebing Wei Xiao Wang Lingxi Li Yisheng Lv The State key laboratory for Management and Control of Complex systems Institute of AutomationChinese Academy of SciencesBeijingChina School of Artificial intelligence University of Chinese Academy of SciencesBeijingChina Electrical and Computer Engineering Indiana University-Purdue University IndianapolisIndianapolisIndianaUSA Complex systems Institute of AutomationChinese Academy of SciencesBeijingChina and School of Artificial IntelligenceUniversity of Chinese Academy of SciencesBeijingChina

Purpose–The purpose of this paper aims to model interaction relationship of traffic agents for motion prediction,which is critical for autonomous *** is obvious that traffic agents’trajectories are influenced by physical lane rules and agents’social ***/methodology/approach–In this paper,the authors propose the social relation and physical lane aggregator for multimodal motion prediction,where the social relations of agents are mainly captured with graph convolutional networks and self-attention mechanism and then fused with the physical lane via the self-attention ***–The proposed methods are evaluated on the Waymo Open Motion Dataset,and the results show the effectiveness of the proposed two feature aggregation modules for trajectory ***/value–This paper proposes a new design method to extract traffic interactions,and the attention mechanism is used in each part of the model to extract and fuse different relational features,which is different from other methods and improves the accuracy of the LSTM-based trajectory prediction method.

关键词： Deep learning Machine learning Autonomous driving Trajectory prediction

Memristor-based multi-synaptic spiking neuron circuit for spiking neural network

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

Chinese Physics B 2022年第4期31卷 225-233页

作者： Wenwu Jiang Jie Li Hongbo Liu Xicong Qian Yuan Ge Lidan Wang Shukai Duan College of Artificial intelligence Southwest UniversityChongqing 400715China National&Local Joint Engineering laboratory of intelligent Transmission and Control Technology Chongqing 400715China Brain-inspired Computing and intelligent Control of Chongqing key laboratory Chongqing 400715China Chongqing Brain Science Collaborative Innovation Center Chongqing 400715China

Spiking neural networks(SNNs) are widely used in many fields because they work closer to biological ***,due to its computational complexity,many SNNs implementations are limited to computer ***,this paper proposes a multi-synaptic circuit(MSC) based on memristor,which realizes the multi-synapse connection between neurons and the multi-delay transmission of pulse *** synapse circuit participates in the calculation of the network while transmitting the pulse signal,and completes the complex calculations on the software with ***,a new spiking neuron circuit based on the leaky integrate-and-fire(LIF) model is designed in this *** amplitude and width of the pulse emitted by the spiking neuron circuit can be adjusted as *** combination of spiking neuron circuit and MSC forms the multi-synaptic spiking neuron(MSSN).The MSSN was simulated in PSPICE and the expected result was obtained,which verified the feasibility of the ***,a small SNN was designed based on the mathematical model of *** the SNN is trained and optimized,it obtains a good accuracy in the classification of the IRIS-dataset,which verifies the practicability of the design in the network.

关键词： memristor multi-synaptic circuit spiking neuron spiking neural network(SNN)

Robust optimal control for anaerobic-anoxic-oxic reactors

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2021年第7期64卷 1485-1499页

作者： HAN HongGui ZHANG JiaCheng DU ShengLi SUN Hao Yuan QIAO JunFei Faculty of Information Technology Beijing University of TechnologyBeijing100124China beijing key laboratory of computational intelligence and intelligent System Beijing100124China Engineering Research Center of Digital Community Ministry of EducationBeijing100124China

Anaerobic-anoxic-oxic(A_2O) reactors, as the core parts of wastewater treatment process(WWTP), have attracted considerable attention to achieve the reliability of denitrification and dephosphorization. However, it is difficult to realize the optimal operation of A_2O reactors due to the existence of nonlinear dynamics and large uncertainties. To solve this problem, a robust optimal control(ROC) strategy is developed to improve the operation performance of A_2O reactors. First, data-driven systematic evaluation criteria are developed to describe the operational indicators of changeable conditions. Second, a robust optimization algorithm is designed to select the optimal solution. Third, a fuzzy neural network(FNN) is used to track the optimal solution in the control process. Finally, this proposed ROC strategy is applied to the phosphorus removal benchmark simulation model(BSM1-P) and the real A_2O reactors. The results demonstrate that the strategy developed in this paper has great potential for application in real A_2O reactors.

关键词： anaerobic-anoxic-oxic reactors robust optimal control robust optimization fuzzy neural network

A spintronic memristive circuit on the optimized RBF-MLP neural network

在线全文

学校读者我要写书评

暂无评论

Chinese Physics B 2022年第11期31卷 272-283页

作者： Yuan Ge Jie Li Wenwu Jiang Lidan Wang Shukai Duan School of Artificial intelligence Southwest UniversityChongqing 400715China Chongqing Brain Science Collaborative Innovation Center Chongqing 400715China Brain-inspired Computing and intelligent Control of Chongqing key laboratory Chongqing 400715China National&Local Joint Engineering laboratory of intelligent Transmission and Control Technology Chongqing 400715China

A radial basis function network(RBF)has excellent generalization ability and approximation accuracy when its parameters are set ***,when relying only on traditional methods,it is difficult to obtain optimal network parameters and construct a stable model as *** view of this,a novel radial basis neural network(RBF-MLP)is proposed in this *** connecting two networks to work cooperatively,the RBF’s parameters can be adjusted adaptively by the structure of the multi-layer perceptron(MLP)to realize the effect of the backpropagation updating ***,a genetic algorithm is used to optimize the network’s hidden layer to confirm the optimal neurons(basis function)number *** addition,a memristive circuit model is proposed to realize the neural network’s operation based on the characteristics of spin *** is verified that the network can adaptively construct a network model with outstanding robustness and can stably achieve 98.33%accuracy in the processing of the Modified National Institute of Standards and Technology(MNIST)dataset classification *** experimental results show that the method has considerable application value.

关键词： radial basis function network(RBF) genetic algorithm spintronic memristor memristive circuit

SpikeGoogle:Spiking Neural Networks with GoogLeNet-like inception module

在线全文

学校读者我要写书评

暂无评论

CAAI Transactions on intelligence Technology 2022年第3期7卷 492-502页

作者： Xuan Wang Minghong Zhong Hoiyuen Cheng Junjie Xie Yingchu Zhou Jun Ren Mengyuan Liu School of intelligent systems Engineering Sun Yat-sen UniversityShenzhenChinaGuangdong Provincial Key Laboratory of Fire Science and Intelligent Emergency TechnologyGuangzhouChina Shenzhen Academy of Metrology and Quality Inspection ShenzhenChina Infocare systems Limited New Zealand

Spiking Neural Network is known as the third-generation artificial neural network whose development has great *** the help of Spike Layer Error Reassignment in Time for error back-propagation,this work presents a new network called SpikeGoogle,which is implemented with GoogLeNet-like inception *** this inception module,different convolution kernels and max-pooling layer are included to capture deep features across diverse *** results on small NMNIST dataset verify the results of the authors’proposed SpikeGoogle,which outperforms the previous Spiking Convolutional Neural Network method by a large margin.

关键词： GoogLeNet inception Spiking Neural Networks

Monolayer molecular crystals for low-energy consumption optical synaptic transistors

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

Nano Research 2022年第8期15卷 7639-7645页

作者： Zhekun Hua Ben Yang Junyao Zhang Dandan Hao Pu Guo Jie Liu Lang Jiang Jia Huang Interdisciplinary Materials Research Center School of Materials Science and EngineeringFrontiers Science Center for Intelligent Autonomous SystemsTongji UniversityShanghai 201804China beijing National laboratory for Molecular Sciences Key Laboratory of Organic SolidsInstitute of ChemistryChinese Academy of SciencesBeijing 100190China Translational Research Institute of Brain and Brain-Like intelligence Shanghai Fourth People's Hospital Affiliated to Tongji UniversitySchool of Materials Science and EngineeringTongji UniversityShanghai 200434China

Artificial synaptic devices hold great potential in building neuromorphic *** to the unique morphological features,twodimensional organic semiconductors at the monolayer limit show interesting properties when acting as the active layers for organic field-effect ***,organic synaptic transistors are prepared with 1,4-bis((5’-hexyl-2,2’-bithiophen-5-yl)ethyl)benzene(HTEB)monolayer molecular *** similar to biological synapses,including excitatory postsynaptic current(EPSC),pair-pulse facilitation,and short/long-term memory,have been *** synaptic device achieves the minimum power consumption of 4.29 fJ at low drain voltage of−0.01 ***,the HTEB synaptic device exhibits excellent long-term memory with 109 s EPSC estimated retention ***-like functions such as dynamic learning-forgetting process and visual noise reduction are demonstrated by nine *** unique morphological features of the monolayer molecular semiconductors help to reveal the device working mechanism,and the synaptic behaviors of the devices can be attributed to oxygen induced energy *** work shows the potential of artificial neuroelectronic devices based on organic monolayer molecular crystals.

关键词： monolayer molecular crystals synaptic devices organic field-effect transistors optical synaptic transistors