检索结果-南通市图书馆

Representation learning via a semi-supervised stacked distance autoencoder for image classification

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2020年第7期21卷 1005-1018页

作者： Liang HOU Xiao-yi LUO zi-yang wang Jun LIANG College of Control Science and Engineering Zhejiang UniversityHangzhou 310027China

Image classification is an important application of deep learning. In a typical classification task, the classification accuracy is strongly related to the features that are extracted via deep learning methods. An autoencoder is a special type of neural network, often used for dimensionality reduction and feature extraction. The proposed method is based on the traditional autoencoder, incorporating the "distance" information between samples from different categories. The model is called a semisupervised distance autoencoder. Each layer is first pre-trained in an unsupervised manner. In the subsequent supervised training, the optimized parameters are set as the initial values. To obtain more suitable features, we use a stacked model to replace the basic autoencoder structure with a single hidden layer. A series of experiments are carried out to test the performance of different models on several datasets, including the MNIST dataset, street view house numbers(SVHN) dataset, German traffic sign recognition benchmark(GTSRB), and CIFAR-10 dataset. The proposed semi-supervised distance autoencoder method is compared with the traditional autoencoder, sparse autoencoder, and supervised autoencoder. Experimental results verify the effectiveness of the proposed model.

关键词： Autoencoder Image classification Semi-supervised learning Neural network

An Immersive Virtual Reality System for Rodents in Behavioral and Neural Research

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5期18卷 838-848页

作者： Li Liu zi-yang wang Yu Liu Chun Xu Institute of Automation Chinese Academy of SciencesBeijing 100190China Shanghai Center for Brain Science and Brain-inspired Intelligence Technology Shanghai 200031China Institute of Neuroscience Chinese Academy of SciencesShanghai 200031China

Context cognition involves abstractly deriving meaning from situational information in the world and is an important psychological function of higher cognition. However, due to the complexity of contextual information processing, along with the lack of relevant technical tools, little remains known about the neural mechanisms and behavioral regulation of context cognition. At present, behavioral training with rodents using virtual reality techniques is considered a potential key for uncovering the neurobiological mechanisms of context cognition. Although virtual reality technology has been preliminarily applied in the study of context cognition in recent years, there remains a lack of virtual scenario integration of multi-sensory information, along with a need for convenient experimental design platforms for researchers who have little programming experience. Therefore, in order to solve problems related to the authenticity, immersion, interaction, and flexibility of rodent virtual reality systems, an immersive virtual reality system based on visual programming was constructed in this study. The system had the ability to flexibly modulate rodent interactive 3 D dynamic experimental environments. The system included a central control unit, virtual perception unit, virtual motion unit, virtual vision unit, and video recording unit. The neural circuit mechanisms in various environments could be effectively studied by combining two-photon imaging and other neural activity recording methods. In addition, to verify the proposed system′s performance, licking experiments were conducted with experimental mice. The results demonstrated that the system could provide a new method and tool for analyzing the neural circuits of the higher cognitive functions in rodents.

关键词： Virtual space flexible control multi-sensory interactions visual programming context cognition

Accelerated haze removal for a single image by dark channel prior

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2019年第8期20卷 1109-1119页

作者： Bo-xuan YUE Kang-ling LIU zi-yang wang Jun LIANG State Key Laboratory of Industrial Control Technology Zhejiang University

Haze scatters light transmitted in the air and reduces the visibility of *** with haze is still a challenge for image processing applications *** the purpose of haze removal,we propose an accelerated dehazing method based on single *** other methods based on regions,our method estimates the transmission map and atmospheric light for each pixel independently,so that all parameters can be evaluated in one traverse,which is a key to ***,the transmission map is bilaterally filtered to restore the relationship between *** restoration via the linear hazy model,the restored images are tuned to improve the contrast,value,and saturation,in particular to offset the intensity errors in different channels caused by the corresponding *** experimental results demonstrate that the proposed dehazing method outperforms the state-of-the-art dehazing methods in terms of processing *** with other dehazing methods and quantitative criteria(peak signal-to-noise ratio,detectable marginal rate,and information entropy difference)are introduced to verify its performance.

关键词： Haze removal Dark channel prior Hazy image model Bilateral filtering

Some notes on numerical simulation of the turbulent cavitating flow with a dynamic cubic nonlinear sub-grid scale model in OpenFOAM

在线全文

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2020年第4期32卷 790-794页

作者： zi-yang wang Xian-bei Huang Huai-yu Cheng Bin Ji State Key Laboratory of Water Resources and Hydropower Engineering Science School of Power and Mechanical EngineeringWuhan UniversityWuhan 430072China College of Electrical Energy and Power EngineeringYangzhou UniversityYangzhou 225127China

The accuracy of large eddy simulation(LES)is highly dependent on the performance of sub-grid scale(SGS)*** the present paper,a dynamic cubic nonlinear sub-grid scale model(DCNM)proposed by Huang et *** implemented for the simulation of unsteady cavitating flow around a 3-D Clark-Y hydrofoil in *** performance in predicting the evolution of cloud cavitation is discussed in *** simulation with a linear model,the dynamic Smagorinsky model(DSM),is also conducted as a *** results with DCNM show a better agreement with the available experimental *** comparison between DCNM and DSM further suggests that the DCNM is able to predict the backscatter more precisely,which is an important feature in *** characteristics of DCNM is analyzed to account for its advantages in the prediction of unsteady cloud cavitation as *** results reveal that it is the nonlinear terms of DCNM that makes DCNM capture sub-grid scale vortices better and more suitable for studying the transient behaviors of cloud cavitation than DSM.

关键词： Cavitation large eddy simulation(LES) nonlinear sub-grid scale model OpenFOAM

Numerical simulation of multi-scale cavitating flow with special emphasis on the influence of vortex on micro-bubbles

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Hydrodynamics 2022年第6期34卷 1032-1043页

作者： Ming-zhe Zhou zi-yang wang Xiao-rui Bai Huai-yu Cheng Bin Ji State Key Laboratory of Water Resources and Hydropower Engineering Science Wuhan UniversityWuhan430072China

A multi-scale Euler-Lagrange method is applied in the current paper to investigate the characteristics of turbulent cloud cavitating flow around a Clark-Y hydrofoil,in which macroscopic cavitating structures are simulated by volume of fluid(VOF)approach,while micro-scale bubbles are modelled based on Rayleigh-Plesset equation and bubble motion *** numerical results are in reasonable agreement with the available experiments,and the transition between multi-scale structures is captured *** evolution of microscopic bubble behaviors is statistically *** one typical cycle,numbers and Sauter mean radius of the bubbles show a similar variation tendency which both sharply spike to their maxima after the breakup of the sheet cavity,and then decrease continuously to their minima before the next *** bubbles are mainly concentrated at the tail of the attached cavity,in front of the cloud cavity and in the region quite close to the hydrofoil suction ***,vortices are extracted to account for the micro-scale *** is found that intense vortices aggravate turbulence fluctuation,thus spalling cavity to generate massive micro ***,the vortices provide microscopic bubbles low pressure and detain them so that numerous bubbles are able to *** effects are remarkable after the breakup of attached sheet cavity,owing to the violent vortices *** the contrary,when sheet cavity develops,the flow field becomes comparatively stable since the strong vortices travel *** micro-scale bubbles are produced,and their volume is generally small.

关键词： Cloud cavitation Euler-Lagrange method bubble-vortex interaction bubble dynamics

Development of a portable multi-channel system for plant physiological signal recording

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Information Processing in Agriculture 2016年第2期3卷 124-132页

作者： Ting Li zi-yang wang Dong-Jie Zhao Lan Huang Zhong-Yi wang College of Information and Electrical Engineering China Agricultural UniversityBeijing 100083China Modern Precision Agriculture System Integration Research Key Laboratory of Ministry of Education Beijing 100083China Key Laboratory of Agricultural Information Acquisition Technology(Beijing) Ministry of AgricultureBeijing 100083China

Bioelectrical signals can reflect physiological state of organs or tissues in plants and have a significant potential value in research of plant stress *** order to study the relationship between environment factors and electrical signals in plant,a portable multichannel physiological signal acquisition system which relevant in plant physiology research was *** parameters and electrical signals can be measured in different channels by the acquisition system simultaneously and the measurement data will be displayed in an embedded integrated touch screen which is the system processing *** system was validated to be stable and reliable after the calibration and repeated experiments of recording electrical signals in Helianthus annuus L.

关键词： Bioelectrical signals Acquisition system Multi-channel Portable device

Semi-supervised remote sensing image scene classification with prototype-based consistency

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 Journal of Aeronautics 2024年第2期37卷 459-470页

作者： yang LI Zhang LI zi wang Kun wang Qifeng YU College of Aerospace Science and Engineering National University of Defense TechnologyChangsha 410073China

Deep learning significantly improves the accuracy of remote sensing image scene classification,benefiting from the large-scale ***,annotating the remote sensing images is time-consuming and even tough for *** neural networks trained using a few labeled samples usually generalize less to new unseen *** this paper,we propose a semi-supervised approach for remote sensing image scene classification based on the prototype-based consistency,by exploring massive unlabeled *** this end,we,first,propose a feature enhancement module to extract discriminative *** is achieved by focusing the model on the foreground ***,the prototype-based classifier is introduced to the framework,which is used to acquire consistent feature *** conduct a series of experiments on NWPU-RESISC45 and Aerial Image Dataset(AID).Our method improves the State-Of-The-Art(SOTA)method on NWPU-RESISC45 from 92.03%to 93.08%and on AID from 94.25%to 95.24%in terms of accuracy.

关键词： Semi-supervised learning Remote sensing Scene classification Prototype network Deep learning

Photocatalytic activity of ferric oxide/titanium dioxide nanocomposite films on stainless steel fabricated by anodization and ion implantation

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 2013年第8期20卷 725-732页

作者： Wei-ting Zhan Hong-wei Ni Rong-sheng Chen Gao Yue Jun-kai Tai zi-yang wang Key Laboratory for Ferrous Metallurgy and Resources Utilization of the Ministry of Education Wuhan University of Science and Technology

A simple surface treatment was used to develop photocatalytic activity for stainless steel. AISI 304 stainless steel specimens after anodization were implanted by Ti ions at an extracting voltage of 50 kV with an implantation dose of 3 × 10^15 ***-2 and then annealed in air at 450℃ for 2 h. The morphology was observed by scanning electron microscopy. The microstructure was characterized by X-ray diffraction and X-ray photoelectron spectroscopy. The photocatalytic degradation of methylene blue solution was carried out under ultraviolet light. The corrosion resistance of the stainless steel was evaluated in NaCl solution （3.5 wt%） by electrochemical polarization curves. It is found that the Ti ions depth profile resembles a Gaussian distribution in the implanted layer. The nanostructured Fe203/TiO2 composite film exhibits a remarkable enhancement in photocatalytic activity referenced to the mechanically polished specimen and anodized specimen. Meanwhile, the annealed Ti-implanted specimen remains good corrosion resistance.

关键词： surface treatment stainless steel nanocomposite films photocatalysis ion implantation electrochemicalanodization

The calibration model in potassium ion flux non-invasive measurement of plants in vivo in situ

在线全文

学校读者我要写书评

暂无评论

Information Processing in Agriculture 2016年第2期3卷 76-82页

作者： Lin Xue Dong-jie Zhao zi-yang wang Xiao-dong wang Cheng wang Lan Huang Zhong-yi wang College of Information and Electrical Engineering China Agricultural UniversityBeijing 100083China College of Information Technology Beijing Union UniversityBeijing 100101China Modern Precision Agriculture System Integration Research Key Laboratory of Ministry of Education Beijing 100083China Key Laboratory of Agricultural Information Acquisition Technology(Beijing) Ministry of AgricultureBeijing 100083China National Engineering Research Center for Information Technology in Agriculture Beijing 100097China

SIET(Self-referencing Ion Electrode Technique)provides a novel electrophysiological tool which can non-invasively measure the dynamic influxes and effluxes of ions caused by the diffusion along the concentration gradients in ***,in this technique ion fluxes are converted to voltage signals using an ion selective microelectrode at a small amplitude of lV,which is easy to be interfered by the ambient ***,effective solutions to the suppression of noise and calibration of ion flux measurement system are very important for this method.A K^+-selective microelectrode was constructed using liquid ion exchangers(LIX)to investigate ion transport over plant tissue.A standard concentration gradient which simulates plant living cells was produced by an electrode with a certain tip diameter,filled with a solution containing a known K+concentration in 100 mmol/*** ion diffusion simulation model was *** model evaluated the performance of ion flux measurement system in accuracy and reliability by comparing the consistency of the measured value and the predicted curve.K^+fluxes were measured within 25 min at each measuring point of distance 10,20,30,40,50,80,and 100 lm from the K+source,*** can be seen that the K^+fluxes changes little,which indicates that ion flux measurement system has a reliable *** study provides a theoretical basis for a new non-invasive ion flux measurement method creation and a new sensors design.

关键词： SIET Plant K^+flux Calibration In situ