检索结果-南通市图书馆

Point-voxel dual transformer for LiDAR 3D object detection

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Optoelectronics Letters 2025年

作者： Tong Jigang yang Fanhang yang sen Du Shengzhi Tianjin Key Laboratory for Control Theory & Applications in Complicated Systems and Intelligent Robot La-boratory， Tianjin University of Technology， 391 Binshui West Road Department of Electrical Engineering， Tshwane University of Technology

Deep learning for point clouds faces the challenge of their inherent unordered nature， which makes traditional CNN-like methods not directly applicable. However， due to the inherent permutation invariance， transformers provide solutions to unordered points problems faced in LiDAR-based object detection. In this paper， a two-stage LiDAR 3D object detection framework is presented， namely Point-Voxel Dual Transformer (PV-DT3D)， which is a transformer-based method. In the proposed PV-DT3D， point-voxel fusion features are used for proposal refinement. Specifically， in the PV-DT3D， keypoints are sampled from entire point cloud scene and used to encode representative scene features via a proposal-aware voxel set abstraction module. Subsequently， following the generation of proposals by the region proposal networks (RPN)， the internal encoded keypoints are fed into dual transformer encoder-decoder architecture. In 3D object detection， the proposed PV-DT3D is the first to take advantage of both pointwise transformer and channel-wise architecture to capture contextual information from the perspective of spatial and channel dimensions. Experiments conducted on the highly competitive KITTI 3D car detection leaderboard show that， the PV-DT3D achieves superior detection accuracy among state-of-the-art point-voxel-based methods.

关键词：

Hyperbolic cosine transformer for LiDAR 3D object detection

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Optoelectronics Letters 2025年

作者： Tong Jigang yang Fanhang yang sen Du Shengzhi Tianjin Key Laboratory for Control Theory & Applications in Complicated Systems and Intelligent Robot Laboratory， Tianjin University of Technology Department of Electrical Engineering， Tshwane University of Technology

Recently， Transformer has achieved great success in computer vision. However， it is constrained because the spatial and temporal complexity grows quadratically with the number of large points in 3D object detection applications from point clouds. Previous point-wise methods are suffering from time consumption and limited receptive fields to capture information among points. To address these limitations， we propose the cosh-attention， which reduces the computation complexity of space and time from the quadratic order to linear order with respect to the number of points. In the cosh-attention， the traditional softmax operator is replaced by non-negative ReLU activation and hyperbolic-cosine-based operator with re-weighting mechanism. Then based on the key component， cosh-attention， we present a two-stage hyperbolic cosine transformer (ChTR3D) for 3D object detection from point clouds. It refines proposals by applying cosh-attention in linear computation complexity to encode rich contextual relationships among points. Extensive experiments on the widely used KITTI dataset and Waymo Open Dataset demonstrate that， compared with vanilla attention， the cosh-attention significantly improves the inference speed with competitive performance. Experiment results show that， among two-stage state-of-the-art methods using point-level features to refine proposals， the proposed ChTR3D is the fastest one.

关键词：

基于Faster R-CNN和Mask R-CNN的滑坡自动识别研究

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

大地测量与地球动力学 2025年第1期45卷 1-4,12页

作者：于宪煜杨森湖北工业大学土木建筑与环境学院武汉市430068

基于高分一号影像,以三峡库区库首段为例,通过目视解译出160个滑坡样本,按照9∶1比例分为训练样本和验证样本,分别利用Faster R-CNN和Mask R-CNN算法构建滑坡自动识别模型。为进一步对比分析不同样本比例下两种模型的性能,分别采用8∶2... 详细信息

基于高分一号影像,以三峡库区库首段为例,通过目视解译出160个滑坡样本,按照9∶1比例分为训练样本和验证样本,分别利用Faster R-CNN和Mask R-CNN算法构建滑坡自动识别模型。为进一步对比分析不同样本比例下两种模型的性能,分别采用8∶2、7∶3、6∶4的样本比例进行计算。研究结果表明,Mask R-CNN模型识别结果准确率、召回率和F 1分数等3项指标均优于Faster R-CNN;且经过交叉验证,证明Mask R-CNN模型的性能更为稳定。

关键词：深度学习滑坡识别 Mask R-CNN Faster R-CNN 交叉验证

Experimental study on the space charge properties in haze events

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2020年第1期32卷 361-376页

作者： Xiaobin Zhang sen yang Tianli Bo Key Laboratory of Mechanics on Western Disaster and Environment Lanzhou University

In recent years,haze has posed a serious threat to the global climate change,ecological balance and human *** this study,the laboratory experiments and field observations were performed and a possible charging mechanism was proposed to investigate the space charge properties in haze *** laboratory experiments showed that the charge polarity of primary aerosol is determined by species of combustion fuels while the magnitude is dependent on the combustion *** field observations revealed that the space charge of atmosphere aerosol in haze events differs from that of fair weather and is closely related to PM2.5 concentration when Relative Humidity(RH)<60%,with 1 to 2 orders of magnitude less than the case when RH≥60%.The analysis of equivalent chargeto-mass ratio(ECTM)suggested that in haze events the space charge is governed by primary aerosol emitted by combustion of fossil fuel in a low relative humidity,whereas it is manipulated by the secondary chemical reaction of atmosphere aerosol in a high relative *** we can identify the main pollutants in haze events according to the polarity of atmosphere aerosol and quickly take measures when RH<60%.Accordingly,the dusthaze of RH<80% can be divided into dry-dust-haze when RH<60%and wet-dust-haze when 60%≤RH<80%.Our study firstly elucidated the space charge properties of atmosphere aerosol in haze events and can provide a new perspective for the prevention and control of air pollution.

关键词： Haze Atmosphere aerosol Space charge density(SCD) Equivalent charge-to-mass ratio(ECTM) Charging mechanisms

An extended state observer with adjustable bandwidth for measurement noise

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Engineering and Electronics 2024年第1期35卷 233-241页

作者： ZHANG Shihua QI Xiaohui yang sen Shijiazhuang Campus Army Engineering University of PLAShijiazhuang 050053China

In this paper,a bandwidth-adjustable extended state observer(ABESO)is proposed for the systems with measurement *** is known that increasing the bandwidth of the observer improves the tracking speed but tolerates noise,which conflicts with observation ***,we introduce a bandwidth scaling factor such that ABESO is formulated to a 2-degree-of-freedom *** observer gain is determined and the bandwidth scaling factor adjusts the bandwidth according to the tracking *** the tracking error decreases,the bandwidth decreases to suppress the noise,otherwise the bandwidth does not *** is proven that the error dynamics are bounded and converge in finite *** relationship between the upper bound of the estimation error and the scaling factor is *** the scaling factor is less than 1,the ABESO has higher estimation accuracy than the linear extended state observer(LESO).Simulations of an uncertain nonlinear system with compound disturbances show that the proposed ABESO can successfully estimate the total disturbance in noisy *** mean error of total disturbance of ABESO is 15.28% lower than that of LESO.

关键词： extended state observer(ESO) boundedness and convergence adjustable bandwidth measurement noise

Optofluidic transport and assembly of nanoparticles using an all-dielectric quasi-BIC metasurface

在线全文

学校读者我要写书评

暂无评论

Light(Science & Applications) 2023年第9期12卷 1771-1781页

作者： sen yang Justus C.Ndukaife Department of Electrical and Computer Engineering Vanderbilt UniversityNashvilleTNUSA Interdisciplinary Materials Science Vanderbilt UniversityNashvilleTNUSA Full list of author information is available at the end of the article

Manipulating fluids by light at the micro/nanoscale has been a long-sought-after goal for lab-on-a-chip *** heating has been demonstrated to control microfluidic dynamics due to the enhanced and confined light absorption from the intrinsic losses of ***,the counterpart of metals,has been used to avoid undesired thermal effects due to its negligible light ***,we report an innovative optofluidic system that leverages a quasi-BIC-driven all-dielectric metasurface to achieve subwavelength scale control of temperature and fluid *** experiments show that suspended particles down to 200 nanometers can be rapidly aggregated to the center of the illuminated metasurface with a velocity of tens of micrometers per second,and up to millimeter-scale particle transport is *** strong electromagnetic field enhancement of the quasi-BIC resonance increases the flow velocity up to three times compared with the off-resonant situation by tuning the wavelength within several nanometers *** also experimentally investigate the dynamics of particle aggregation with respect to laser wavelength and power.A physical model is presented and simulated to elucidate the phenomena and surfactants are added to the nanoparticle colloid to validate the *** study demonstrates the application of the recently emerged all-dielectric thermonanophotonics in dealing with functional liquids and opens new frontiers in harnessing non-plasmonic nanophotonics to manipulate microfluidic ***,the synergistic effects of optofluidics and high-Q all-dielectric nanostructures hold enormous potential in high-sensitivity biosensing applications.

关键词： dielectric quasi negligible

In-situ TiC Reinforced Composite Coating Produced by Powder Feeding Laser Cladding

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

Journal of Materials Science & Technology 2006年第4期22卷 519-525页

作者： sen yang Wenjin LIU Minlin ZHONG Department of Materials Science and Engineering Inner Mongolia University of Technology Hohhot 010062 China Department of Mechanical Engineering Tsinghua University Beijing 100084 China

A Ni-base alloy composite coating reinforced with TiC particles of various shapes and sizes on medium carbon steel substrate was produced by multilayer laser cladding. The chemical compositions, microstructures and surface morphology of the cladded layer were analyzed using energy dispersive X-ray spectroscopy （EDX）, scanning electron microscope （SEM）, and X-ray diffractometry （XRD）. The experimental results showed that an excellent metallurgical bonding between the coating and the substrate was obtained. The microstructure of the coating was mainly composed of γ-Ni dendrites, a small amount of CrB, Ni3B, M23C6 and dispersed TiC particles. Much more and larger TiC particles formed in the overlapping zone, which led to a slightly higher microhardness of this zone. The maximum microhardness of the coating was about HV0.21200. The effects of the laser processing parameters on the microstructures and properties of coating were also investigated.

关键词： Laser cladding In-situ synthesized Powder feeding TiC particle

Lanthanide-DNA supramolecular hydrogels with tunable and responsive luminescence

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2022年第5期65卷 1043-1051页

作者： yang sen PAN XiaoRui TANG JianPu YAO Chi yang DaYong Frontiers Science Center for Synthetic Biology Key Laboratory of Systems Bioengineering(MOE)Institute of Biomolecular and Biomedical EngineeringSchool of Chemical Engineering and TechnologyTianjin UniversityTianjin300350China

Supramolecular DNA hydrogels have been synthesized based on the assembly of DNA building-blocks such as branched DNA and long DNA *** structures and functions of sole-module DNA hydrogels remain *** methodologies by integrating hybrid components are desired to expand the synthesis of DNA ***,we synthesized a Ln^(3+)-containing luminescent supramolecular hydrogel by employing the coordination and electrostatic interactions between lanthanide ions(Tb^(3+)and Eu^(3+))and linear single-stranded DNA(ssDNA).Through the coordination between ssDNA and Ln^(3+),a series of luminescent supramolecular hydrogels were synthesized,among which the Tb-G_(n)/T_(n)and Eu-T_(n)hydrogels emitted the characteristic luminescence of Tb and Eu,*** luminescent intensities of the hydrogels were adjusted by designing DNA sequences with programmable bases and chain ***,the Tb/Eu co-doped luminescent supramolecular hydrogel displayed tunable luminescence from green to yellow by regulating the stoichiometric ratio of Tb/***,the hydrogel had reversible luminescent stimulation responsiveness toward Ag^(+)/*** expected that the synthesis of Ln^(3+)-containing luminescent supramolecular hydrogels enriched the strategies of the construction of DNA hydrogels,and promoted the development of stimuli-responsive supramolecular materials.

关键词： DNA hydrogel supramolecular self-assembly luminescent hydrogel lanthanide complex stimuli-responsiveness

Aerodynamic-Parameter Identification and Attitude Control of Quad-Rotor Model with CIFER and Adaptive LADRC

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Mechanical Engineering 2021年第2期34卷 150-159页

作者： sen yang Leiping Xi Jiaxing Hao Wenjie Wang Department of UAV Engineering Army Engineering UniversityShijiazhuang 050003China School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100083China School of Electrical and Electronic Engineering Shijiazhuang Tiedao UniversityShijiazhuang 050043China School of Aerospace Engineering Beijing Institute of TechnologyBeijing 100081China

Current research on quadrotor modeling mainly focuses on theoretical analysis methods and experimental methods,which have problems such as weak adaptability to the environment,high test costs,and long ***,the PID controller,which is currently widely used in quadrotors,requires improvement in ***,the aforementioned research has considerable practical significance for the modeling and controller design of quadrotors with strong coupling and nonlinear *** the present research,an aerodynamic-parameter estimation method and an adaptive attitude control method based on the linear active disturbance rejection controller(LADRC)are designed ***,the motion model,dynamics model,and control allocation model of the quad-rotor are established according to the aerodynamic theory and Newton-Euler ***,a more accurate attitude model of the quad-rotor is obtained by using a tool called CIFER to identify the aerodynamic parameters with large uncertainties in the frequency ***,an adaptive attitude decoupling controller based on the LADRC is designed to solve the problem of the poor anti-interference ability of the quad-rotor and adjust the key control parameter b0 automatically according to the change in the moment of inertia in real ***,the proposed approach is verified on a semi-physical simulation platform,and it increases the tracking speed and accuracy of the controller,as well as the anti-disturbance performance and robustness of the control *** paper proposes an effective aerodynamic-parameter identification method using CIFER and an adaptive attitude decoupling controller with a sufficient anti-interference ability.

关键词： Quad-rotor Qarameters identification CIFER Adaptive LADRC